Describe a goal and coordinated agents plan it, run the steps, and deliver the
result. Every sensitive action waits at the approvals gate and is audit-logged.
Bring your own model — your keys stay encrypted.
Goal
“Summarise last week's pipeline and email the team”
Plan
Project kickoff, phase-by-phase plan
Agent
Runs each step with its tools, skills & powers
Approval
Outbound email pauses for a human ✓
Output
Delivered & logged for audit
Built for the work your team already does
Stand up automation your teams can rely on, across the functions that move fastest with it.
RevOps
Turn pipeline reviews and account research into scheduled, agent-run routines.
Finance
Reconcile, summarise and report on a schedule — with approvals before anything sends.
IT
Automate onboarding, access requests and routine tickets end to end.
Support
Draft, route and resolve — keeping a human in the loop on every customer-facing step.
Build AI agents you can actually trust
Connect any model. Approve every outbound action. Keep a full audit trail.
Bring your own model
Google Gemini by default, with OpenAI and Anthropic provider slots. Your API key is
encrypted at rest and scoped to your account.
Human-in-the-loop approvals
Outbound actions — send email, create an event, call an API — pause at the Approvals
gate until you approve them. Nothing leaves without a sign-off.
Audit everything
Every sensitive action is logged for review. Export or delete your data
from Settings at any time.
Visual when you want it. Code when you need it.
Workflow Automation gives you a drag-and-drop canvas — build a flow by connecting
simple steps, with human approval on anything sensitive.
Build flows on a visual canvas, or let AI draft one from a prompt.
Runs reliably on its own — steps retry and pick up where they left off.
Connect the tools you already use.
Start from a template, or import a workflow you already have.
// a step, anywhere in your flowreturn items.map(i => ({
json: {
account: i.json.name,
risk: scoreLead(i.json),
// pauses here if risk is high
approval: i.json.amount > 10000,
},
}));
One platform, six modules
Independent, but wired into the same AI agents and approvals.
Work Management
Turn goals into tracked, assignable tasks — organised into projects and sprints, for people and AI agents.
Workflow Automation
Build automations on a visual canvas, or from a plain-English prompt, with human approval on anything sensitive.
Customer Management (CRM)
Accounts and contacts, pipeline and deals, leads, activities and a dashboard — all in one place.
Marketing
Plan and run campaigns across content, social, email, ads and web — with a marketing copilot and per-discipline AI agents, every send approval-gated.
Workforce Management
Manage people and AI agents as one team: roles, accountability, skills and reviews.
Finance
Keep proper books: accounts, invoices, bills and clear financial reports.
How the modules work together
See how a single process moves across the platform — with AI agents and people working side by side.
1
Customer Management
captures the new lead with its account context.
2
Workflow Automation
triggers the right automation the moment it arrives.
3
AI Agent
researches the company and surfaces what matters.
4
Work Management
turns the findings into assigned, tracked tasks.
5
Workforce Management
records accountability and routes approvals to a person.
6
Finance
raises the invoice and keeps the books when the deal closes.
The result: agents and people run the same business process together — with approvals and an audit trail wherever it counts.
Enterprise-ready controls
Reliable. Auditable. Yours to control.
Security & access
Five-tier IAM roles (viewer → owner) with per-project overrides. BYO API keys,
encrypted at rest.
Observability
Audit logs for every sensitive action, usage metering and a public system-status
signal.
Workspace structure
Organizations → Projects, each with its own agents, tools, skills and powers.
AI governance
Human-in-the-loop approvals, guardrails and per-power permission policies.
Describe a goal — your agents plan, execute, and deliver it.
Ask SMAIVIZ
Suggested questions
We use cookies — essential ones to run SMAIVIZ, and optional ones for preferences
and anonymous diagnostics. You choose what we may store; nothing optional is set
until you allow it. See our
Cookie Policy.
Cookie preferences
Choose which cookies SMAIVIZ may use. Necessary cookies are always on; everything
else stays off until you allow it. Full details in the
Cookie Policy.
Appearance
Strictly necessaryAlways on
Required for sign-in, security (CSRF protection), remembering your cookie choice, and your appearance & accessibility settings. The service can't run without these, so they can't be switched off.
Functional
Remembers your language and chat-panel state across visits. Stored locally on your device.
Analytics
Anonymous, scrubbed front-end error diagnostics so we can find and fix bugs. No third-party analytics and no profiling.
Marketing
We don't use advertising or third-party tracking cookies today. This stays off unless that ever changes.
Choose a new password
Delete account
Your account will be scheduled for deletion. We keep your data for
90 days so you can sign in again to reactivate. All
sessions and API keys are revoked immediately. After 90 days the
account is permanently removed and audit-log entries are anonymized.
Please review our Terms & Conditions
and Privacy Policy and accept to continue.
We log security events for governance review; you can export or delete your data anytime from Settings.
No recent chats yet.
No plans yet — describe a one-off task in the planner above and the engine will build a plan for it.
No routines yet — a Routine runs Agents on a schedule (or on-demand) and produces an Output.
No outputs yet — Routines produce these when they run.
Loading history…
Section
Frequently asked questions
Do I need to sign in to try SMAIVIZ?
You can browse the public pages — overview, documentation, FAQs, security, terms,
privacy and the blog — without an account. To use the app (chat, projects,
automations and your own AI assistants), you sign in so your work stays private to you.
Do I need to know how to code?
No. You describe what you want in plain English and SMAIVIZ helps build it — whether
that's an automation, an AI assistant, or a report.
What can I do with SMAIVIZ?
Run your business from one place: manage customers and deals, organise your team of
people and AI agents, plan and track work, automate the busywork, and build AI
assistants that do real tasks for you.
Which AI models can I use?
You can choose the AI model that best fits each task, and bring your own model key if
you prefer — so you stay in control of model choice, data, and cost.
Can AI act on its own, or do I stay in control?
You stay in control. Anything sensitive waits for your approval before it happens, and
you decide exactly what each AI agent is allowed to do.
Can I work as a team?
Yes. Invite your teammates and choose what each person can see and do — from
view-only access up to full control.
Is my data private and secure?
Yes. Your data and any keys you add are kept private to your account and encrypted, and
only you and the people you invite can see your workspace. You can also keep your data
in the region you choose.
Can I export or delete my data?
Anytime. Export everything we hold about you in one file from your settings, and delete
your account whenever you like — with a short grace window in case you change your mind.
How much does it cost?
You can start for free and upgrade when you're ready, so you only pay as your needs
grow.
Build an AI agent and chat with it or run it on a schedule.
Run your sales pipeline and keep customer records in one place.
Manage and govern a mixed human + AI workforce.
Automate repetitive processes end to end.
Agents & automation
Build AI workers, give them capabilities, and run them on a schedule or on
demand — with a human in the loop for anything sensitive.
Highlights
Agents — configurable AI workers you chat with or run.
Skills, Powers & Tools — ready-made abilities you add to an agent so it can do more.
Workflows — turn a plain-language goal into a reviewable, phase-by-phase plan.
Routines — scheduled or on-demand jobs that run agents and deliver an output.
Approvals — sensitive or outbound actions pause for human sign-off.
Top use cases
Stand up a personal or team assistant in minutes.
Automate a recurring report or task on a schedule.
Scope a goal into a plan before any work runs.
Keep AI actions safe with approval gates.
Kriya — work management
Plan, assign and track work across your team — with people and AI
agents as assignees.
Highlights
Tasks with owners, status, due dates, comments, files and dependencies.
Assign any task to a teammate or an AI agent that can do it.
Projects, sections and sprints to structure larger efforts.
Cadence to repeat work, plus AI-suggested tasks.
Top use cases
Run team projects and sprints in one tracker.
Hand routine tasks to AI agents.
Automate recurring work and follow-ups.
Hive — workforce
Manage your whole workforce — people and AI agents — as one
organization.
Highlights
One directory of people and agents, with onboarding and offboarding.
Positions, requisitions and an org chart that survives turnover.
Keep AI accountable: every agent answers to a person, with limits you set.
Approvals, performance reviews, capacity and cost in one place.
Top use cases
Run HR for a blended human + AI team.
Govern what each AI agent can do and who's responsible.
Plan capacity and understand the cost of work.
Orbit — CRM
Your CRM — the companies you serve, the people inside them, and the deals in
your pipeline.
Highlights
Accounts and contacts with full activity timelines.
A drag-to-advance sales pipeline with forecasting.
Leads that convert to accounts, contacts and deals in one click.
A dashboard of pipeline value, win rate and team performance.
Top use cases
Track and manage your sales pipeline.
Keep a single source of truth for customer relationships.
Qualify leads and convert them without re-typing.
Echo — marketing
Plan and run all your marketing in one place — content & SEO, social,
email & SMS, ads and web — with a marketing copilot and a team of per-discipline AI
agents, and a human approving anything that sends, spends or publishes.
Highlights
A marketing cockpit and a unified calendar across every channel.
Campaigns that group content, social posts, email and ads under one goal and budget.
Content & SEO pipeline, social scheduling, email/SMS broadcasts and lifecycle journeys.
Ad spend, pacing and ROAS, plus landing pages, popups and forms.
A/B experiments, multi-touch attribution, referrals and events.
Plan a campaign from a goal and run it across every channel.
Draft on-brand content, posts and emails with AI — approve before anything goes out.
Track spend, ROAS and which channels actually drive results.
Grow with referrals, events and landing pages that convert.
Flux — workflow automation
Build automated workflows visually and run them across your tools — with AI
help.
Highlights
Build automations visually by connecting simple steps.
Describe what you want in plain English and get a draft to start from.
Run them on a schedule, or whenever something happens.
Add human approval steps, and see a clear history of every run.
Top use cases
Automate multi-step processes across apps.
Add governed AI steps into existing workflows.
Replace manual, repetitive ops work.
Artha — finance
Keep proper books and see where your money goes — accounts, invoices, bills and clear reports.
Highlights
Track income and expenses with proper books.
Send invoices, and record bills and payments.
See clear financial reports at any time.
Work in more than one currency.
Top use cases
Keep your accounts in order without a separate tool.
Invoice customers and track what you're owed.
Understand profit, cash and spend at a glance.
Data & compliance
Keep your data private, in the right place, and easy to account for.
Highlights
Handle requests to access or delete personal data.
Keep data in the region you choose.
Review sensitive data and keep a clear record of important actions.
Log and manage any privacy or security incident.
Top use cases
Respond to data-subject and deletion requests.
Keep data in the right region.
Prove who did what, for audits.
Payment & subscription terms
Version: 2026-06-14 · Last updated: 14 June 2026.
These Payment & Subscription Terms (the “Payment Terms”) govern your purchase
and use of paid plans on SMAIVIZ, operated by Smaiviz Pvt. Ltd.
(“SMAIVIZ”, “we”, “us”). They supplement, and are in
addition to, our Terms of Service and Privacy Policy, which you also accept. By selecting a
plan and clicking I agree, you accept these Payment Terms.
1. Plans, seats & billing cycle
Paid plans are billed per seat, per billing cycle (monthly unless stated
otherwise) in advance. The price shown at checkout for your region applies. We may change
plan pricing or features on a prospective basis with notice; changes do not affect the cycle
you have already paid for.
2. Free trial
Paid plans may start with a free trial of the length shown at checkout
(e.g. 7 days). A valid payment method is required to start the trial. Unless you
cancel before the trial ends, the plan automatically converts to a paid subscription
and we charge your payment method the then-current fee for your selected plan and seats. You
can cancel at any time during the trial from Billing settings to avoid being charged.
3. Automatic renewal & cancellation
Subscriptions renew automatically at the end of each billing cycle until you
cancel. You authorise us (and our payment processors) to store your payment method and charge
it for each renewal. You may cancel renewal at any time from Billing settings; cancellation
takes effect at the end of the current paid cycle, and you retain access until then. We do
not provide pro-rated refunds for partial cycles except where required by law (see §6).
4. Payment processing
Payments are processed by third-party providers — Stripe (for customers
billed in USD) and Razorpay (for customers billed in INR). Your card and
payment details are handled by these processors under their terms and privacy policies; we do
not store full card numbers. You are responsible for keeping a valid payment method on file.
5. Taxes (incl. GST)
Fees are exclusive of taxes unless stated. You are responsible for any applicable sales tax,
VAT, or GST. For Indian customers, 18% GST applies; provide
your GSTIN at checkout to receive a tax invoice. Where we are required to
collect tax, it is added at checkout.
6. Refunds
Except where required by applicable law or expressly stated, fees are non-refundable
and there are no refunds or credits for partial periods, unused seats, or features not used.
If you believe you were charged in error, contact billing support and we will review in good
faith.
7. Failed payments & suspension
If a charge fails, we may retry and notify you. If payment remains unpaid, we may
downgrade, suspend, or restrict access to paid features until the balance is
settled. We are not liable for loss of access arising from a failed payment.
8. Plan changes
You can upgrade, downgrade, or change seat counts from Billing settings. Upgrades may be
charged immediately (prorated by the processor where supported); downgrades take effect at the
next cycle. Enterprise plans are governed by a separate order form or agreement.
9. Changes to these terms
We may update these Payment Terms. Material changes are versioned; when the version changes we
will ask you to review and accept the updated terms before continuing to use paid features.
10. Governing law & contact
These Payment Terms are governed by the laws stated in our Terms of Service. For billing
questions, contact billing@smaiviz.com.
Everything that governs your use of SMAIVIZ, in one place. Pick a
policy from the left, or jump in below.
These policies are published by Smaiviz Private Limited (“SMAIVIZ”),
registered office 7-365, Mahidhara Luxuria, Muthangi, Hyderabad – 502300, Telangana, India.
They should be read together — each cross-references the others where relevant.
Privacy & data
Privacy Policy — what we collect, how we use it, and your rights by region.
Cookie Policy — cookies and similar technologies we use.
A Data Processing Agreement (DPA) is available to business customers on request.
Privacy policy
Effective date: 11 June 2026 · Last updated: 16 June 2026.
This Privacy Policy explains how Smaiviz Pvt. Ltd. (“SMAIVIZ”, “we”,
“us”), a company incorporated in India with its registered office at
7-365, Mahidhara Luxuria, Muthangi, Hyderabad – 502300, India, collects, uses, shares, and
protects personal information, and the choices and rights
you have. Where a specific law gives you additional rights, the
region-specific section (§14) controls for residents of that region.
1. Who we are & our role
For individual (B2C) accounts, SMAIVIZ is the controller /
business / Data Fiduciary of the personal data you
provide. For business (B2B) customers, the customer organisation is the controller of its
end-users' data and SMAIVIZ acts as a processor /
service provider / Data Processor on its documented
instructions. Each organisation manages its controller posture (jurisdiction, DSAR
onboarding, data-protection contact, residency) at
Settings → Data & Compliance.
Contact for privacy matters, and our Grievance Officer (India) /
data-protection contact for other regions, is
privacy@smaiviz.com (see §17). Business-customer
end-users should contact their
organisation's administrator first.
2. Key terms
“Personal data” (also “personal information” in the US, and
“personal data” of a “Data Principal” under India's DPDP Act) means
information that identifies or relates to an identifiable individual. Under India's Digital
Personal Data Protection Act 2023 (“DPDP Act”) you are a Data
Principal and we are a Data Fiduciary. Under US state laws you are
a consumer; under UK GDPR you are a data subject.
“Customer Data” means prompts, files, configurations, workflows, tasks,
outputs, and other content submitted to or generated through the Service.
3. Information we collect
Account data: email address, optional display name, password hash (never plaintext), Google subject identifier when OAuth is used.
Session & device data: session identifier hash, IP address, user-agent, and session creation/expiry timestamps.
Credentials you add: third-party API keys you provide, encrypted at rest. We display only the last four characters in the UI.
Content you create: chats, prompts, agent/routine configurations, files, tasks, and the outputs the Service produces for you.
Audit log: action name, target, IP, user-agent, timestamp, and optional JSON metadata for sensitive actions. We do not write chat or routine contents to the audit log.
Consent records: consent version, acceptance timestamp, and IP — including your cookie-consent receipts (see the Cookie Policy).
Operational metrics: aggregated routine counts, token usage, and durations — not linked to chat contents.
Support communications: messages you send us and our responses.
Sources. We collect this information directly from you, automatically from
your use of the Service (sessions, security logs, diagnostics), and — for B2B accounts —
from the organisation that provisioned your account.
Workforce data. Organizations may use the Service to manage information
relating to workers, including people and AI agents.
Such information is processed solely to provide workforce-management, governance,
accountability, reporting, compliance, and operational functionality requested by the
customer.
Sensitive information. We do not require sensitive data to use the Service.
Input guardrails actively block payment-card and national-ID numbers and
redact emails/phone numbers from prompts. Where an organisation knowingly
processes special categories (children/teen, health, financial/payment, biometric), it must
declare them under Settings → Data & Compliance,
which records the lawful basis and safeguards. We do not use sensitive personal information
to infer characteristics about you.
4. How we use your information
Provide and operate the Service (authentication, sessions, agents, routine execution, delivering outputs).
Secure the Service (rate limiting, intrusion and fraud detection, abuse prevention).
Maintain a governance audit trail for sensitive and outbound actions.
Improve reliability and performance using aggregated, de-identified metrics and (with your consent) scrubbed error diagnostics.
Communicate with you about the Service, security, and support.
Comply with legal obligations and respond to lawful requests.
We do not use your prompts, content, or outputs to train our own models, and we do not sell your personal information.
5. Legal bases & grounds for processing
Where required (UK GDPR and similar), we rely on:
Performance of a contract — to provide the Service you signed up for.
Legitimate interests — security, fraud prevention, and service improvement (balanced against your rights), including:
Governance and accountability controls
Security monitoring and incident response
Audit-trail maintenance and compliance review
Abuse prevention and platform integrity
Consent — for optional cookies/diagnostics and any feature clearly marked as consent-based; you may withdraw consent at any time.
Legal obligation — retention of records as required by law.
Under India's DPDP Act we process personal data on your consent (given
after clear notice and itemised to purpose) or for permitted legitimate uses
(such as a purpose for which you voluntarily provided data, or to comply with law). In the
US, processing is disclosed at or before collection (§14.1) rather than relying on a consent
“legal basis”.
6. Cookies & similar technologies
We use first-party cookies and local storage as described in our
Cookie Policy. Non-essential cookies load only with
your consent; you can change or withdraw your choice any time via
.
7. How we share information
We disclose personal data only to the following categories of recipients, for the purposes shown:
AI model providers (Google Gemini by default; OpenAI, Anthropic, Microsoft Azure OpenAI, Amazon Bedrock, and others you select) — your prompts and outputs are transmitted for inference. When you bring your own key, processing is governed by your agreement with that provider.
Cloud & infrastructure providers (e.g. Google Cloud, AWS, Azure) — to host and run the Service.
Connectors you enable (e.g. email, calendar, Slack, GitHub) — only the data needed for the action you request, after the Approvals gate where applicable.
Professional advisers and authorities — only when required by law and after appropriate review.
A successor in a merger, acquisition, or asset sale, under equivalent protections.
Third-party AI providers may retain, process, or log data according to their own terms,
privacy policies, and operational requirements.
Customers should review the privacy and security commitments of any providers they choose
to use through the Service.
We do not sell personal information, and we do not share it
for cross-context behavioural advertising (as those terms are defined under US state laws).
We do not use third-party advertising networks or advertising cookies.
Operational analytics, monitoring, security, fraud-prevention, and support technologies may
be used to operate, secure, maintain, and improve the Service.
8. AI processing & automated decisions
SMAIVIZ is an AI platform: agents process the prompts and content you give them to plan and
run tasks. Outbound or sensitive actions pause at the Approvals gate for a
human to authorise. We do not make decisions that produce legal or similarly
significant effects based solely on automated processing without meaningful human
involvement, and we do not use your data for profiling for advertising. You can request human
review of any output. AI can produce inaccurate results; do not rely on outputs for legal,
medical, financial, or other professional decisions without independent verification.
8A. AI governance & accountability
SMAIVIZ provides governance controls designed to support responsible deployment of AI
systems. These controls may include:
Human approval workflows
Accountability chains
Agent autonomy controls
Audit logging
Permissions and access controls
Guardrails and policy enforcement
Organizations remain responsible for configuring these controls appropriately for their
use cases. Every AI agent action can be associated with a responsible human through the
accountability framework.
AI-generated outputs may be similar or identical to outputs generated for other users.
SMAIVIZ does not guarantee the uniqueness of AI-generated content.
9. International data transfers
Your data may be processed in regions selected by the deployment operator and the providers
listed in §7, which may be outside your country. Where data leaves your jurisdiction we rely
on appropriate safeguards — for example the EU/UK Standard Contractual
Clauses and the UK International Data Transfer Addendum, Australia's APP 8 reasonable
steps, Singapore PDPA transfer-limitation safeguards, Canadian comparable-protection
measures, and India's DPDP cross-border rules (transfers permitted except to countries the
Government restricts). Business customers can declare a preferred
data-residency region under
Settings → Data & Compliance; this preference is
advisory and actual residency depends on the deployment's infrastructure.
Transfer mechanisms and safeguards may change over time as laws, regulations, regulatory
guidance, and international transfer frameworks evolve.
10. Data retention
Account data: kept until you delete your account, then a 90-day grace window during which you can sign in to reactivate; after that the user row is permanently removed.
Audit log: 365 days by default (configurable). After deletion the entries are anonymised (user_id set to NULL) but the trail remains for compliance review.
Sessions: revoked on logout, password change, or account deletion; otherwise expire after one week of inactivity.
API keys: until you revoke them, or automatically on account deletion.
Cookie-consent receipts: kept to evidence your choice; the matching user link is removed on account deletion.
Backup retention
Deleted information may remain in encrypted backups for a limited period before being
automatically overwritten or removed according to operational backup-retention schedules.
Backup copies are protected using the same security controls applied to active systems
where reasonably practicable.
11. Security
We use strong, memory-hard password hashing, authenticated encryption for secrets at
rest, CSRF protection, rate limiting, secure cookies, brute-force throttling, and audit
logging. See our
Security Policy for additional information
regarding safeguards, governance controls, and security practices. No security
control is absolute; please use a unique password and protect your devices. We will notify
you and the relevant authorities of a personal-data breach as required by law (§15).
11A. Security documentation & audits
Enterprise customers may request reasonable security documentation describing our security
practices. To protect the Service and other customers, intrusive testing — penetration tests,
vulnerability or load testing, source-code review, or similar — requires our prior written
consent. To report a suspected vulnerability, email
security@smaiviz.com.
12. Children's privacy
The Service is not directed to children. We do not knowingly collect data from children
below the age of digital consent in your jurisdiction. If we become aware that a child
below the applicable minimum age has created an account without required parental or
guardian consent, we will take reasonable steps to remove the account and associated
personal data in accordance with applicable law. In the United States we
follow COPPA (under 13) and do not sell or
share the personal information of consumers we know to be under 16 without opt-in. In
India, processing the data of anyone under 18 requires verifiable consent of
a parent or lawful guardian, and we do not undertake tracking, behavioural monitoring, or
targeted advertising toward children.
13. Your privacy rights & how to exercise them
Subject to your region (§14), you may have the right to:
Access & portability — export your data any time as machine-readable JSON from Settings → Privacy & Data.
Correct / rectify — update your profile in Settings.
Delete / erase — delete your account from the Account page; a 90-day grace window applies, after which data is permanently purged (credentials, sessions, consents cascade).
Restrict / object — ask us to freeze processing; while restricted your data is kept and stays accessible/exportable/erasable but is not otherwise actively processed (AI generation is blocked).
Withdraw consent — as easily as you gave it (e.g. ), without affecting prior lawful processing.
Human review — of any AI output (§8).
Complain — to your data-protection regulator (§17).
To make a request, use the in-product tools above or contact us (§17). We verify requests
against your account, respond within the statutory window for your region (commonly ~30 days
for UK GDPR / India DPDP, ~45 days for US state laws, extendable as the law allows), and do
not discriminate against you for exercising your rights. You may use an authorised agent
where the law permits. Business-customer end-users: requests are routed
through the controller's DSAR service at
Settings → Data & Compliance; data a controller
holds outside SMAIVIZ remains its responsibility.
14. Region-specific rights
14.1 United States
If you are a US resident, you have rights under the California Consumer Privacy Act as
amended by the CPRA and under comparable laws in states including Virginia, Colorado,
Connecticut, Utah, Texas, Oregon, Montana, and others. These include the rights to
know/access the categories and specific pieces of personal information we
collect, the sources, business purposes, and categories of recipients; to
delete and correct your information; to opt out of
the sale or sharing of personal information and of targeted advertising; to
limit the use of sensitive personal information; to non-discrimination
for exercising rights; and, in several states, to appeal a denied request.
Categories we collect (CCPA): identifiers (email, IP); internet/network activity (session and security logs); user content you provide; and inferences only as needed to operate the Service.
No sale or sharing. We do not sell or share personal information and have not in the preceding 12 months. Because we do not sell or share, there is nothing to opt out of — but you may still record a preference under Preferences → Privacy, and we honour Global Privacy Control (GPC) signals.
Sensitive personal information. We do not use or disclose it for purposes beyond providing the Service.
Authorised agents may submit requests with proof of authorisation; we may still verify your identity.
“Shine the Light” (California Civil Code §1798.83): we do not disclose personal information to third parties for their direct-marketing purposes.
14.2 India (Digital Personal Data Protection Act, 2023)
If you are in India, you are a Data Principal and we are a Data
Fiduciary. We process your personal data based on your consent
(given after clear notice, itemised to purpose, and withdrawable) or a permitted
legitimate use. You have the right to:
Access a summary of the personal data we process and the processing activities and recipients.
Correction, completion, updating, and erasure of your personal data.
Grievance redressal — raise a grievance with our Grievance Officer (§17), who will respond within the period prescribed by law.
Nominate another individual to exercise your rights in the event of death or incapacity.
Withdraw consent at any time, as easily as it was given.
If your grievance is not resolved, you may escalate to the Data Protection Board of
India. Where you may manage consent through a registered Consent
Manager, we will honour valid instructions received through it. Children's data is
handled per §12 (verifiable parental consent for under-18s; no tracking, behavioural
monitoring, or targeted advertising). We will report personal-data breaches to the Board and
affected Data Principals as required.
14.3 United Kingdom (UK GDPR & Data Protection Act 2018)
You have the rights of access, rectification, erasure, restriction, portability, and to
object to processing, plus rights relating to automated decision-making (§8). Our legal bases
are in §5. You may lodge a complaint with the Information Commissioner's Office
(ICO).
14.4 Canada (PIPEDA & provincial laws)
You may access your personal information, challenge its accuracy, and withdraw consent
(subject to legal or contractual limits). Quebec residents have additional rights under Law
25. You may complain to the Office of the Privacy Commissioner of Canada (OPC)
or your provincial regulator. We report breaches posing a real risk of significant harm to
the OPC and to you.
14.5 Australia (Privacy Act 1988 & the Australian Privacy Principles)
You may request access to and correction of your personal information, and you may deal with
us anonymously or by pseudonym where lawful and practicable. Cross-border disclosures follow
APP 8. Eligible data breaches are notified under the Notifiable Data Breaches scheme. You may
complain to the Office of the Australian Information Commissioner (OAIC).
14.6 Singapore (Personal Data Protection Act)
You may request access to and correction of your personal data, and withdraw consent. We
collect, use, and disclose personal data only for purposes a reasonable person would consider
appropriate and that we have notified to you. Our data-protection contact (DPO) is in §17,
and you may complain to the Personal Data Protection Commission (PDPC). We
notify notifiable breaches to the PDPC and affected individuals.
14.7 All other regions
If you are outside the regions named above, we still apply the protections in this policy as
our baseline. Where your local data-protection law grants you rights — such as to access,
correct, or delete your data, to object to or restrict processing, to withdraw consent, or to
complain to a supervisory authority — we will honour them. Contact us (§17) and we will
respond in line with applicable law.
Subprocessors
We may engage trusted subprocessors and service providers to assist in delivering the
Service. Such subprocessors may provide infrastructure, hosting, security, analytics,
communications, customer support, or AI processing services.
We require subprocessors to maintain appropriate confidentiality, privacy, and security
obligations. A list of significant subprocessors may be published separately or provided
upon request.
15. Data-breach notification
If a breach of security leads to the unauthorised access, loss, or disclosure of your
personal data, we will assess it and notify the competent authority and affected individuals
where required and within the timelines set by your law — for example without undue delay and
within 72 hours to the ICO (UK), to the Data Protection Board of India and affected
principals, to the OAIC (Australia) and PDPC (Singapore) for notifiable breaches, to the OPC
(Canada) for real-risk-of-significant-harm breaches, and as required by applicable US state
breach-notification laws.
16. Changes to this policy
We may update this policy. Material changes will be announced via an in-app banner and the
“Last updated” date above will change. Continued use after the effective date
indicates acceptance of the updated policy.
17. How to contact us & complain
For privacy questions, requests, grievances, or to exercise your privacy rights, contact:
Business-customer end-users should contact their organization's administrator first
regarding data controlled by that organization.
You also have the right to complain to your applicable regulator, including:
Data Protection Board of India (India)
Information Commissioner's Office (United Kingdom)
Office of the Privacy Commissioner of Canada
Office of the Australian Information Commissioner
Personal Data Protection Commission (Singapore)
California Privacy Protection Agency and applicable State Attorneys General (United States)
Cookie policy
Last updated: 11 June 2026 · Policy version:2026-06-11
This Cookie Policy explains how Smaiviz Pvt. Ltd. (“SMAIVIZ”) uses cookies and
similar technologies, and the choices you have. It supplements our
Privacy Policy.
1. What are cookies & similar technologies
Cookies are small files a site stores in your browser. We also use comparable
local-storage entries on your device. Together we call these
“cookies” in this policy. All of the cookies we use are
first-party (set by SMAIVIZ) — we set no third-party advertising
or tracking cookies, and we do not use cookies to sell or share your information or
for cross-context behavioural advertising.
2. Why and on what basis we use cookies
Strictly necessary cookies are used because the Service cannot run without
them (sign-in, security, and remembering your choice); most laws permit these without
consent. For all other cookies our basis depends on your region:
UK, India, Canada, Australia, Singapore — we ask for your prior consent before setting non-essential cookies (UK PECR/UK GDPR; India DPDP; PIPEDA; the Privacy Act/APPs; the PDPA).
United States — we provide notice and choice; you can opt out of non-essential cookies at any time, and we honour Global Privacy Control (GPC) signals where applicable.
3. How we ask for consent
On your first visit we show a cookie banner with equal-prominence Accept all,
Reject non-essential, and Cookie settings choices. Every non-essential
category is off by default; we never load optional cookies before you
choose, and ignoring the banner stores nothing optional. Your choice is recorded against a
policy version, and if we materially change this policy we will ask again. You can change or
withdraw your choice at any time — as easily as you gave it — via the
link in the footer.
4. Categories we use
Strictly necessary — sign-in, security, remembering your cookie choice, and your appearance & accessibility settings (a user-requested preference we keep so the interface stays usable).
Functional — remembers your language and chat-panel state.
Analytics — anonymous, scrubbed front-end error diagnostics so we can fix bugs. No third-party analytics, no profiling.
Marketing — none today. We set no advertising or third-party tracking cookies; this category stays empty unless that ever changes.
5. The cookies we use
All of the following are first-party (provider: SMAIVIZ).
Name
Type
Category
Purpose
Duration
smaiviz_session
Cookie (HttpOnly)
Necessary
Keeps you signed in.
Session lifetime
smaiviz_csrf
Cookie
Necessary
CSRF protection (double-submit token).
Session lifetime
smaiviz_cookie_consent
Cookie
Necessary
Stores your cookie choices and the policy version they apply to.
12 months
smaiviz_authed
Local storage
Necessary
A sign-in breadcrumb that prevents a page flash on load.
Until sign-out / cleared
smaiviz_prefs
Local storage
Necessary
Your appearance & accessibility settings (theme, contrast, motion, text size) — kept as an essential, user-requested preference.
Until cleared
smaiviz_locale
Local storage
Functional
Your chosen language.
Until cleared
chat-dock-state
Local storage
Functional
Remembers whether the chat panel is minimised, open or maximised.
Until cleared
Saved chats cache
Local storage
Functional
A local copy of your recent chats for faster loads.
Until cleared
Anonymous diagnostics
No cookie — network report
Analytics
Sends scrubbed front-end error reports to our servers; nothing is stored in your browser.
n/a
We record a consent receipt (the categories you chose, the policy version,
a timestamp, and your IP address) so we can demonstrate your choice if asked. See the
Privacy Policy for how we handle that data and your
rights.
6. Do Not Track & Global Privacy Control
Because we set no tracking or advertising cookies, there is no cross-site tracking to switch
off. Where applicable in the United States, we treat a recognised Global Privacy
Control (GPC) browser signal as a valid opt-out of any non-essential processing.
7. Managing your choices
Use
to review or change your consent at any time. You can also block or delete cookies through
your browser settings — though strictly-necessary ones will be re-created as needed for the
Service to work, clearing them will sign you out, and disabling local storage may break some
features.
8. Region-specific notes
United States — we provide this notice and honour GPC; we do not sell or share personal information through cookies.
United Kingdom — non-essential cookies are set only with your prior consent under PECR and UK GDPR.
India — non-essential cookies are set only with your consent under the DPDP Act; you may withdraw it at any time.
Canada, Australia, Singapore — we obtain consent for non-essential cookies and limit use to the purposes described above (PIPEDA; the Privacy Act/APPs; the PDPA).
All other regions — where your local law requires consent for non-essential cookies we ask for it; otherwise we apply the above as our baseline and honour any rights your local law provides.
9. Changes to this policy
If we materially change how we use cookies we will bump the policy version, update the date
above, and ask for your consent again the next time you visit.
Version 1.0 · Effective & last updated: 16 June 2026.
This policy describes how SMAIVIZ retains, recovers, deletes, and disposes of Customer
Data, personal data, and system records. Read it together with the
Privacy Policy and the
Subprocessor policy.
Retention principles
We retain information only as long as reasonably necessary for legitimate business
purposes, contractual obligations, legal requirements, security, and operational
continuity — and no longer, unless the law or a contract requires it.
Retention by data type
Customer account data (name, email, account identifiers, billing/subscription info, communication history) — kept while the account is active and for a reasonable period afterwards as needed for legal, operational, and fraud-prevention purposes.
Customer Content (prompts, files, configurations, outputs) — kept while the account is active.
Audit logs — retained for a defined period for security and compliance, then anonymised or removed.
Backups — included in encrypted backups for a limited period before being overwritten or removed on the backup-rotation schedule.
Account termination
Recovery period. After termination, Customer Content may remain recoverable for up to thirty (30) days.
Deletion. Information is then scheduled for deletion from production systems within ninety (90) days.
Retention exceptions
We may retain information longer where necessary for legal/regulatory compliance, fraud
prevention, dispute resolution, security investigations, or where it persists in backups
pending rotation. Anonymised and aggregated information may be retained indefinitely where
permitted by law.
Version 1.0 · Effective & last updated: 16 June 2026.
A “subprocessor” is a third-party service provider engaged by SMAIVIZ that may
process Customer Data in connection with the Service. We select subprocessors for their
security standards, reliability, legal compliance, and privacy protections, and require
them to maintain appropriate confidentiality and security obligations.
AI service providers
OpenAI — AI model processing & generation. Data: prompts, uploaded content, instructions, metadata, generated outputs.
Anthropic — AI model processing & generation. Data: as above.
Google AI services — AI model processing & generation. Data: as above.
Microsoft Azure OpenAI — enterprise AI model services. Data: as above.
Amazon Web Services (Bedrock) — AI model services. Data: as above.
Processing locations: multiple regions as determined by each provider.
When you bring your own key, your use of that provider is governed by your agreement with it.
Infrastructure & operations
Google Cloud — hosting, storage, and infrastructure.
Payment processors (e.g. Stripe, Razorpay) — billing and payment processing.
Analytics, monitoring, communications, and support providers — to operate, secure, and support the Service.
Changes
We may add or replace subprocessors as the Service evolves. An up-to-date list is available
on request at privacy@smaiviz.com; business
customers may request advance notice under their agreement.
Terms & Conditions
Effective date: 16 June 2026 · Last updated: 16 June 2026.
These Terms & Conditions (“Terms”) govern your access to and use of
SMAIVIZ (the “Service”), operated by Smaiviz Pvt. Ltd. (“SMAIVIZ”,
“we”, “us”), a company incorporated in India with its registered
office in Hyderabad, India. Please read them together with our
Privacy Policy and
Cookie Policy.
Definitions
“Service” means the SMAIVIZ platform.
“Account” means a registered user account.
“Organisation” means a legal entity using the Service.
“Workspace” means an environment within the Service.
“User” means any individual accessing the Service.
“Customer Data” means data submitted by users.
“AI Agent” means an automated software agent configured within the Service.
1. Acceptance of these Terms
By creating an account or using the Service, you agree to be bound by these Terms. If you
do not agree, do not register or use the Service. If you use the Service on behalf of an
organisation, you represent that you have authority to bind that organisation, and
“you” refers to that organisation. For business customers, any signed order
form or master agreement and our Data Processing Addendum take precedence over these Terms
to the extent of a conflict.
2. Eligibility
You must be able to form a legally binding contract to use the Service. The Service is
intended for business and professional use and is not directed to children.
Where you are below the age of majority or the age of digital consent in your jurisdiction,
you may use the Service only with the involvement and consent of a parent or legal guardian
to the extent permitted by law. How we handle children's data (including COPPA and the
verifiable parental-consent rules of India's DPDP Act) is described in the
Privacy Policy. If you believe a minor is using
the Service without proper consent, contact
privacy@smaiviz.com.
3. Accounts & security
You are responsible for all activity under your account and for keeping your password and API keys confidential.
Provide accurate registration information and keep it current.
Notify us promptly of any unauthorised access or security incident.
We may suspend or terminate accounts that violate these Terms, abuse the Service, or pose security or legal risk.
4. The Service
SMAIVIZ is a platform for building and running AI agents, workflows, and routines, with
governance, approvals, and an audit trail. We may add, change, or discontinue features. We
will give reasonable notice of materially adverse changes where practical.
The Service includes governance mechanisms such as approvals, accountability chains, audit
logs, autonomy controls, and human-oversight features designed to support the responsible
deployment of AI systems.
Beta features
We may provide features designated as alpha, beta, preview, early access, or experimental.
Such features may be modified, suspended, or withdrawn at any time and are provided without
any service-level commitment or warranty.
Service limits
The Service may impose limits on usage, storage, API calls, workflow executions, agent
runs, file uploads, and similar resources. Exceeding these limits may result in throttling,
additional charges, suspension, or upgrade requirements.
5. Your content
“Your Content” means the inputs you submit (prompts, files, configurations) and
the outputs generated for you. As between you and us, you own Your Content,
subject to the terms of any model or third-party provider you use. You grant us a limited,
worldwide, non-exclusive licence to host, process, and transmit Your Content solely to
operate, secure, and support the Service. We do not use Your Content to train our
own models, and we do not sell it. You are responsible for Your Content and for
having the rights and permissions needed to submit it. Except for Your Content, no
ownership rights in the Service, models, prompts, workflows, templates, or platform
materials are transferred to you.
6. Acceptable use
You agree not to use the Service to:
Violate any law or regulation, or infringe any third party's rights (including IP, privacy, and publicity rights).
Generate or distribute unlawful, harmful, infringing, defamatory, deceptive, or harassing content, malware, or spam.
Attempt to identify individuals from, or build profiles for, unlawful surveillance or discrimination.
Reverse engineer, decompile, or extract source code beyond rights expressly granted, or circumvent security or usage limits.
Interfere with or disrupt the integrity, security, or performance of the Service.
Copy, replicate, or commercially exploit substantial parts of the Service in order to create a competing product, or resell the Service, without our prior written agreement.
Access, scrape, harvest, index, or download the Service or its content by automated means (including bots, crawlers, scrapers, or headless browsers), other than a search engine that honours our robots.txt; or cache, store, or redistribute our content beyond ordinary personal use.
Copy, mirror, frame, or clone the Service's pages, design, layout, text, code, or other content, in whole or in part; or remove, alter, or obscure any proprietary notice or embedded identifier.
You must also comply with the acceptable-use and safety policies of any model or
infrastructure provider you use through the Service.
Your compliance responsibilities
You are responsible for ensuring that your use of the Service complies with applicable
laws, including employment, privacy, tax, export, consumer-protection, and AI-related
regulations.
7. AI outputs
The Service orchestrates third-party AI models. AI output may be inaccurate, incomplete,
biased, or unsuitable for your purpose, and similar prompts can produce different results.
You are responsible for reviewing and validating any output before relying on it,
and you must not rely on it for legal, medical, financial, safety-critical, or other
professional decisions without independent verification. The Service provides assistive
output; a human remains responsible for decisions and for actions taken through approvals.
AI outputs may be substantially similar to content generated for other users, and SMAIVIZ
does not guarantee the uniqueness of generated output.
8. Third-party services & your API keys
Your use of model and cloud providers (e.g. Google Gemini, OpenAI, Anthropic, AWS, Google
Cloud, Microsoft Azure) and of connectors you enable is subject to those providers' own
terms. When you bring your own API key, your use of that provider is governed by your
agreement with it; you authorise us to transmit your prompts and outputs to it to provide
the Service. We do not control third-party providers and are not responsible for their
availability, output, or acts.
9. Intellectual property & feedback
We and our licensors retain all rights in the platform, including its software, design,
and documentation. Except for the rights expressly granted, no licence is implied. If you
send us feedback or suggestions, you grant us a perpetual, irrevocable, royalty-free
licence to use them without restriction.
Open-source software
Certain components may include open-source software governed by their respective licences.
Nothing in these Terms overrides those licences.
10. Fees, billing & taxes
Paid plans are billed in advance through our billing provider; usage beyond plan limits may
be billed separately. Fees are exclusive of taxes, which are your responsibility. Unless
required by law or stated otherwise, fees are non-refundable. We may change pricing with
reasonable advance notice effective from your next billing period.
11. Privacy & data protection
Our handling of personal data is described in the
Privacy Policy. For business customers where we
act as a processor, a Data Processing Addendum is available on request and, once entered
into, forms part of these Terms and governs that processing.
Confidentiality
Each party agrees to protect confidential information received from the other party and to
use it only for purposes related to the Service.
12. Service availability & support
We aim for high availability but do not guarantee uninterrupted or error-free service.
Maintenance and incidents are surfaced via the in-app status indicator; scheduled
maintenance is announced in advance where reasonably practical.
13. Suspension & termination
You may stop using the Service and delete your account at any time from
Settings → Account; a 90-day grace and deletion
process applies as described in the Privacy Policy. We may suspend or terminate access for
breach of these Terms, suspected unlawful or harmful use, or to protect the Service or other
users, with notice where reasonable. Sections intended to survive termination (including
§§5, 7, 9, 14–18, 20, the Confidentiality clause, and the Definitions) continue to apply.
14. Disclaimers
Except as expressly stated and to the extent permitted by law, the Service is provided
“as is” and “as available” without warranties of any kind, express
or implied, including merchantability, fitness for a particular purpose, accuracy, and
non-infringement. Nothing in these Terms excludes or limits any warranty, guarantee,
or liability that cannot be excluded or limited under the law that applies to you —
including statutory consumer guarantees in jurisdictions such as the UK, Australia, India,
Canada, and Singapore.
15. Limitation of liability
To the maximum extent permitted by law, SMAIVIZ and its affiliates will not be liable for
any indirect, incidental, special, consequential, or punitive damages, or any loss of
profits, revenue, data, or goodwill, arising from or related to the Service. To the maximum
extent permitted by law, our total aggregate liability arising out of or related to the
Service will not exceed the greater of the amounts you paid us for the Service in the 12
months before the event giving rise to the claim, or USD 100. These limits do not apply to
liability that cannot be limited by law (such as for death or personal injury caused by
negligence, fraud, or wilful misconduct), and nothing here affects your non-excludable
statutory rights as a consumer.
16. Indemnification
You will indemnify and hold harmless SMAIVIZ and its affiliates from third-party claims,
losses, and reasonable costs arising out of Your Content, your use of the Service, or your
breach of these Terms — except to the extent caused by our own breach or unlawful conduct.
17. Export controls & sanctions
You represent that you are not located in, and will not use the Service in or for the
benefit of, any country or party subject to applicable trade sanctions or export-control
restrictions, and that you will comply with applicable export-control and sanctions laws.
18. Governing law & disputes
These Terms are governed by the laws of India, without regard to conflict-of-laws rules, and
the courts at Hyderabad, Telangana, India have jurisdiction over disputes. This does not
deprive you of the protection of mandatory consumer-protection laws of
your country of residence, or of your right to bring a complaint before, or seek remedies
from, a competent regulator or court where the law gives you that right.
19. Changes to these Terms
We may update these Terms. Material changes will be communicated via in-app banner or
email, and the “Last updated” date above will change. Continued use after the
changes take effect constitutes acceptance; if you do not agree, stop using the Service.
20. General
Entire agreement. These Terms, with the Privacy and Cookie Policies and any order form or DPA, are the entire agreement between you and us about the Service.
Severability. If any provision is unenforceable, the rest remains in effect.
No waiver. Failing to enforce a provision is not a waiver of it.
Assignment. You may not assign these Terms without our consent; we may assign them in connection with a merger, acquisition, or sale of assets.
Force majeure. Neither party is liable for delays or failures caused by events beyond its reasonable control.
Notices. We may give notice via the Service or by email; notices to us go to the address in §21.
No third-party beneficiaries except as expressly stated.
21. Contact
Questions about these Terms can go to
terms@smaiviz.com (or, for organisation accounts,
your workspace administrator). Privacy questions go to
privacy@smaiviz.com.
Version 1.0 · Effective & last updated: 16 June 2026.
This Acceptable Use Policy (“AUP”) governs use of all SMAIVIZ products, APIs,
AI systems, agents, automations, and related services, and forms part of the
Terms. You remain solely responsible for the content,
prompts, workflows, automations, and outputs in your account. Violations may result in
suspension, termination, credit forfeiture, legal action, or reporting to authorities.
You must not use the Service to
Break the law — facilitate crime, fraud, money laundering, or violate sanctions/export-control laws.
Deceive — impersonate people or organisations, forge documents, phish, or run scams or social-engineering.
Attack systems — create or distribute malware/ransomware/spyware, gain unauthorised access, steal credentials, bypass security, or run denial-of-service attacks. (Authorised, lawful defensive security is permitted.)
Generate harmful content — threats, harassment, hateful or discriminatory content, violent-extremist or terrorist material, or unlawful exploitation/abuse material.
Misuse synthetic media — deceptive deepfakes, unauthorised impersonation, or non-consensual synthetic content. (Clearly-disclosed lawful parody, satire, artistic, or educational use may be permitted.)
Infringe IP or privacy — violate copyrights/trademarks/trade secrets or proprietary notices; unlawfully collect, disclose, or surveil personal data.
Run prohibited high-risk use cases — life-critical, emergency, military, weapons, classified, regulated healthcare diagnosis, regulated financial advice, nuclear, or aviation-safety systems, unless expressly authorised in a written Enterprise Agreement.
Abuse the AI/platform — extract model parameters, reverse-engineer systems, circumvent safeguards, or build competing models through unauthorised extraction.
Abuse resources or payments — overload infrastructure, evade usage limits, use stolen payment methods, or commit chargeback/refund/credit fraud.
Resell without authorisation — resell, sublicense, or operate a service bureau on the Service without our prior written agreement.
Automation & agents
You remain responsible for everything your AI agents, automated workflows, scheduled
actions, and integrations do. Delegating an action to an agent does not remove your
responsibility for it.
Enforcement & reporting
We may warn, remove content, restrict, suspend, terminate, forfeit credits, report to
authorities, or take legal action — and may act immediately where needed to protect users,
systems, or rights. Report suspected violations to
support@smaiviz.com.
Refund & Cancellation Policy
Version 1.0 · Effective & last updated: 16 June 2026.
This policy governs subscriptions, prepaid AI credits, renewals, cancellations, refunds,
and chargebacks. It forms part of the Terms. By
purchasing or using the Service, you agree to it.
Prepaid AI credits
Prepaid model. AI usage runs on prepaid credits — you buy credits in advance and consume them through usage. No plan includes a monthly credit allowance.
Free starter credits. Each account receives a one-time grant of 5,000 free credits to try the Service; once used, you top up to continue.
Top-ups. You purchase additional credits in packs. Credits are not currency, have no cash value, and are not redeemable for cash or transferable.
Subscriptions
Paid plans (monthly, annual, or enterprise) are billed in advance and unlock features and
limits — separately from the prepaid credits that power AI usage. You can disable
auto-renewal at any time before the next renewal date; cancellation stops future charges
but does not end the current paid term, which runs to its end.
Refund eligibility
Except as stated here or required by law, purchases are final.
Consumed credits (used to generate outputs) are not refundable. Where only part of a purchase has been used, only the unused portion may be considered, at our discretion.
Generally non-refundable: consumed or expired credits, completed AI processing, generated outputs, and professional/consulting/support services already provided.
Duplicate or unauthorised charges and billing errors (reported within 30 days) are reviewed and adjusted where verified.
Cancellation, chargebacks & outages
You may cancel at any time; access may continue through the paid term. Please contact us
before initiating a chargeback — fraudulent chargebacks may lead to suspension or legal
action. Service availability is governed by any applicable Service Level Agreement; where
it applies, service credits are the sole remedy for qualifying downtime (no automatic cash
refunds). Nothing here limits non-waivable consumer-protection rights.
Requests
Submit refund requests with your account and transaction details to
support@smaiviz.com.
Copyright & DMCA Policy
Version 1.0 · Effective & last updated: 16 June 2026.
SMAIVIZ respects intellectual property rights and expects users to do the same. You may
only upload or process content you own or are authorised to use. This policy sets out how
to report infringement and how we respond.
Reporting infringement
Copyright owners (or their authorised agents) may send a notice including: identification
of the work and the allegedly infringing material; its location; your contact details; a
good-faith statement that the use is unauthorised; a statement under penalty of perjury
that the notice is accurate and you are authorised to act; and your physical or electronic
signature.
On a valid notice we may investigate, request more information, remove or disable access to
content, or restrict accounts. A user who believes content was removed in error may submit
a counter-notification (identifying the material and its prior location, a statement under
penalty of perjury that removal was a mistake, contact details, consent to jurisdiction
where required, and a signature). We may restore content where appropriate.
Repeat infringers, trademarks & trade secrets
We may suspend or terminate accounts associated with repeated violations, and we accept
trademark and trade-secret complaints with supporting detail. Knowingly false reports may
be rejected and may lead to account action or legal remedies.
AI-generated content
AI outputs may be influenced by prompts and training methods. We do not guarantee that
generated outputs are unique or free of third-party claims; you remain responsible for
evaluating IP risk in your use of outputs.
AI Output & Automation Disclaimer
Version 1.0 · Effective & last updated: 16 June 2026.
This Disclaimer governs use of our AI services, agents, automations, and APIs, and forms
part of the Terms. Using the Service, you accept the
risks described here.
AI is probabilistic
AI systems may generate unexpected, inaccurate, incomplete, contradictory, outdated, or
fabricated information (“hallucinations”), and may misinterpret instructions.
These are inherent limitations of current AI. We do not warrant that
outputs are accurate, complete, current, error-free, reliable, or fit for your purpose.
Your responsibility
You are responsible for evaluating, validating, and verifying outputs before relying on
them, and for the business, operational, hiring, and other decisions you make based on
them. Do not rely on AI outputs alone.
Not professional advice
AI outputs do not constitute legal, medical, financial, investment, accounting,
engineering, tax, or other professional advice. Consult qualified professionals where
appropriate, and keep meaningful human oversight for important, legal,
health, financial, employment, and compliance decisions.
Automation, agents & generated code
Automations and AI agents may take incorrect actions, act on incomplete information, or
produce unintended outcomes; you are responsible for monitoring and supervising them, and
for all actions performed by agents you configure or operate. AI-generated code may contain
bugs, security vulnerabilities, or infringing material — review, test, and secure it before
deployment.
Third-party AI providers & regulated uses
Outputs may be affected by third-party providers (OpenAI, Anthropic, Google, Azure OpenAI,
AWS Bedrock, and others) updating or changing their models or services. Unless expressly
agreed in an Enterprise Agreement, the Service is not intended for regulated healthcare,
life-critical, military, weapons, aviation-safety, or classified uses. To the maximum
extent permitted by law, SMAIVIZ is not liable for losses arising from outputs,
hallucinations, automation failures, agent actions, or reliance on generated content.
Responsible AI
Version 1.0 · Effective & last updated: 16 June 2026.
We are committed to designing, deploying, and operating AI in a way that promotes trust,
safety, fairness, accountability, transparency, privacy, security, and lawful use. AI
creates real opportunities alongside risks that must be managed responsibly.
Core principles
Human oversight. Important decisions should stay subject to human judgment; AI assists, it doesn't replace responsible decision-making. Sensitive actions pause at the Approvals gate.
Fairness & non-discrimination. We work to evaluate and reduce unfair bias, though it cannot always be eliminated — review outputs before important decisions.
Transparency. We communicate AI capabilities and limitations, including that outputs may be inaccurate, may hallucinate, and may need verification.
Accountability. We maintain internal responsibility for AI governance, risk, and security controls; every agent action can be traced to a responsible human through the accountability framework.
Privacy & security. We protect data per the Privacy Policy and our security practices, and we do not train our own models on your content.
Reliability & compliance. We monitor AI systems and work to meet applicable legal and regulatory requirements.
Our internal AI governance program (system inventory, risk classification, provider
review, and human-oversight controls) underpins these commitments. Customers remain
responsible for their own use of AI outputs and decisions based on them.
Security & trust
Last updated: 16 June 2026.
Security is foundational to a platform that runs autonomous AI agents on your
behalf. This page summarises how we protect your account, your data, and the actions
taken in your workspace. For how we handle personal data, see our
Privacy Policy and
Cookie Policy.
AI governance & human oversight
SMAIVIZ includes governance controls designed for the responsible deployment and
operation of AI systems. These controls include:
Human approval workflows
Accountability chains
Agent autonomy controls
Audit logging
Permission boundaries
Role-based access controls
Organizations remain responsible for determining appropriate approval requirements,
governance policies, and autonomy settings for their agents.
Account & access
Modern password protection. Passwords are stored using a memory-hard, salted hashing algorithm — never in plain text — and are re-hardened as our parameters evolve.
Single sign-on. Optional Google sign-in (OpenID Connect) lets your team use existing, centrally-managed identities.
Secure sessions. Sign-in uses server-side sessions with hardened, same-site cookies; sessions can be revoked and expire automatically.
Role-based access & approvals. Each workspace has role-based access so people only see what they should, and sensitive agent actions can require explicit human approval.
Encryption
In transit. All traffic is served over HTTPS/TLS.
At rest. Sensitive secrets such as your provider API keys are encrypted with authenticated encryption before storage, held in plain text only for the moment a request needs them, and never written to logs.
Minimal exposure. Once saved, secrets are never returned to the browser — the interface shows only the last few characters so you can recognise a key without revealing it.
Application integrity
Database access is fully parameterised through an ORM, and user-supplied content is escaped before display — guarding against injection and cross-site scripting.
State-changing requests are protected against cross-site request forgery.
Authentication is rate-limited to slow brute-force and credential-stuffing attempts.
An append-only audit trail records security-relevant events, with retention and routine pruning.
We do not include third-party advertising networks in the product experience. Operational analytics, monitoring, security, support, and performance-improvement tooling may be used to operate, secure, maintain, and improve the Service.
Vulnerability management
We monitor security advisories affecting our infrastructure, software components, and
technology stack. Security updates are evaluated and applied according to risk,
severity, and operational impact. Critical vulnerabilities receive priority review and
remediation.
Abuse & bot defence
We apply layered defences at the network edge and the application boundary —
including a web application firewall, optional CAPTCHA, and adaptive rate limits — to
keep automated abuse away from your workspace. Suspicious activity is surfaced to
operators for review.
Monitoring & response
Security-relevant events are logged centrally and monitored, with automated alerts on
anomalous patterns so our team can respond quickly. We maintain operational runbooks
for incident handling.
Your data & AI
You own your content. Your prompts, files, and outputs are yours; we process them only to operate the Service.
No training on your content. We do not use your content to train our own models, and we do not sell it — see our Terms.
Bring your own keys. When you connect a model or cloud provider, your use of that provider is governed by your agreement with it.
Model providers
When AI models are used through the Service, prompts, files, and outputs may be
transmitted to the selected model provider to generate responses. Each provider
processes data according to its own terms, privacy commitments, and security practices.
Customers are responsible for reviewing the terms and policies of any third-party
providers they choose to use.
AI agent security
AI agents operate within permissions, guardrails, limitations, and autonomy levels
configured by the customer. Actions may be restricted, approved, audited, delayed, or
blocked based on organizational policies and governance requirements.
Every agent action can be traced through the accountability framework to a responsible
human. Organizations are responsible for reviewing agent configurations and ensuring
that autonomy settings are appropriate for their intended use cases.
Infrastructure
SMAIVIZ runs on Google Cloud, using managed, regularly-patched services. We restrict
production access on a need-to-know basis and keep configuration secrets separate from
application code.
Backup & recovery
Critical platform data is backed up regularly. Backups are encrypted and retained
according to operational requirements. Recovery procedures are periodically tested to
support restoration in the event of operational incidents, service disruption, or data
loss.
Personnel access
Access to production systems and customer data is restricted to authorized personnel
with a legitimate business need. Access permissions are reviewed periodically and
removed when no longer required. Administrative actions are logged and monitored.
Data residency
Customers may specify a preferred data residency region where supported. Actual storage
and processing locations depend on the deployment infrastructure, service configuration,
and applicable legal requirements.
Authentication roadmap
We continue to enhance account-security capabilities and identity-management features as
the platform evolves. Security features may be expanded over time to include additional
authentication, authorization, monitoring, and access-control capabilities.
Compliance & roadmap
We align our controls with widely-recognised security practices and are progressively
maturing toward formal third-party assessment as we move to general availability. If you
have specific compliance requirements, contact
sales@smaiviz.com.
Please do not publicly disclose vulnerabilities until we have had a reasonable
opportunity to investigate and remediate the issue.
Responsible disclosure
If you believe you've found a security vulnerability, please email
security@smaiviz.com with the details and
steps to reproduce. We welcome good-faith research: we will acknowledge your report and
will not pursue or support legal action against researchers who act in good faith, avoid
privacy violations and service disruption, and give us reasonable time to remediate
before any public disclosure.
Information Security
Version 1.0 · Effective & last updated: 16 June 2026.
This summarises the security principles and controls we use to protect Customer Data,
systems, and infrastructure. For the customer-facing overview see also
Security & trust.
Objectives
Our security program supports confidentiality, integrity, availability, accountability,
resilience, and regulatory compliance, and is reviewed and updated periodically.
Controls
Access control. Least-privilege, need-to-know access; uniquely-assigned accounts; periodic review; prompt revocation when no longer required.
Encryption. Data encrypted in transit (HTTPS/TLS) and sensitive secrets encrypted at rest with authenticated encryption.
Monitoring & logging. Centralised logging of security-relevant events with alerting on anomalies.
Vulnerability management. We monitor advisories and apply updates by risk and severity.
Resilience. Encrypted, monitored, periodically-tested backups and multi-region cloud infrastructure for continuity and recovery.
Shared responsibility
You remain responsible for protecting your credentials, securing your endpoints, and
managing user access within your organisation. No control is absolute.
Incident Response
Version 1.0 · Effective & last updated: 16 June 2026.
We maintain a framework to identify, assess, contain, investigate, remediate, and
communicate security incidents affecting our systems or your data.
What counts as an incident
An actual or suspected event that may affect the confidentiality, integrity, availability,
or security of systems or data — for example unauthorised access, credential compromise,
malware/ransomware, data exposure, denial-of-service, or insider threats.
How we respond
Detect & classify incidents by severity.
Contain & investigate with a designated response team (coordination, investigation, remediation, communications, legal review).
Remediate & recover services and preserve evidence.
Notify. Where required by law or contract, we notify affected customers and authorities within a commercially reasonable period, and cooperate on investigation and mitigation.
Improve. We review incidents to strengthen future security.
See the Privacy Policy (data-breach notification)
for how we handle personal-data breaches.
Vulnerability Disclosure Policy
Version 1.0 · Effective & last updated: 16 June 2026.
We welcome good-faith security research that helps us protect customers. This policy
explains how to report a vulnerability and what's in and out of scope.
Scope
www.smaiviz.com and SMAIVIZ web applications, APIs, AI services, cloud services, and hosted
infrastructure we own and operate.
Out of scope
Denial-of-service / DDoS, ransomware or malware deployment, social engineering or phishing
of our people, physical security testing, credential theft, account-takeover attempts,
spam, and any destruction, alteration, or unauthorised access of customer data.
Good-faith research
Avoid privacy violations and service disruption, don't access more data than necessary,
don't modify customer information, and don't maintain persistent access.
How to report
Email security@smaiviz.com (subject: “Security
Vulnerability Report”) with a description, affected systems/URLs, reproduction steps,
a proof of concept where appropriate, a severity assessment, and your contact details.
Safe harbour
If you act in good faith, follow this policy, avoid prohibited activities, report promptly,
and don't exploit a vulnerability beyond what's needed to demonstrate it, we will not pursue
or support legal action against you and will work with you on remediation.
The SMAIVIZ blog
Real stories about running more of your business with AI — automating the busywork,
keeping every customer in one place, managing people and AI agents as one team, keeping
the books, and staying in control while AI does more. Plain, practical ideas you can put
to work today.
Promoted by SMAIVIZ — smaiviz.com.
Automate the busywork: workflows that run themselves
It's a quarter to six on a Friday. The office has mostly emptied out, and you're still at your desk, building the same report you built last Friday, and the Friday before that. Pull the numbers from one place. Paste them into another. Reformat the columns the way your manager likes. Write the little summary at the top that nobody reads until the one week you skip it. You know this routine so well you could do it half-asleep, which is roughly how you're doing it now. Somewhere across town, someone you care about is wondering when you'll be home.
One repetitive loop, redrawn as a flow that runs on its own — with a checkpoint before anything leaves the building.
Here's the thing about busywork: it doesn't feel like much in the moment. Twenty minutes here, ten minutes there. The welcome email to the new client. The follow-up nudge to the three people who went quiet. Copying yesterday's orders from one tool into the spreadsheet that feeds the other tool. None of it is hard. None of it is even interesting. That's exactly why it's so corrosive.
Because it adds up, quietly, in a currency you can't get back. It's the reason you're never quite off — your mind keeps a background tab open, running through the list of things that only happen if you personally make them happen. It's the small thing that slips when you're swamped: the renewal nobody flagged, the new customer who never got their welcome note and felt, on day one, a little forgotten. The work is invisible right up until the moment it's missed, and then it's suddenly very visible, with your name attached.
The cost isn't the minutes. It's the vigilance.
If someone asked you to add up the hours, you probably couldn't. The damage isn't really measured in time. It's measured in attention — in the fact that a part of your brain is permanently on guard duty, making sure the routine things actually get done. That vigilance is what follows you to dinner. It's what wakes you at 5 a.m. with the thought did I send that?
And it scales the wrong way. The more your business grows, the more of these little loops there are, and the more of them route through you specifically. Success doesn't lighten the load; it multiplies it. The reward for handling the busywork well is more busywork.
So let me ask the question you've probably been avoiding: which of the tasks you'll do this week actually need you — your judgment, your relationships, your taste — and which ones just need someone, reliably, on time, forever?
Be honest. It's a longer list than you'd like. The Friday report needs the same numbers pulled the same way every time. The welcome email needs to go out whenever a new client is added. The quiet-deal nudge needs to fire when a deal hasn't moved in two weeks. None of those need your brain. They need a pattern, followed faithfully. That's not a job for a person. That's a job for a process that simply runs.
What if you could just describe it?
This is where SMAIVIZ comes in, and specifically a part of it called Flux. Think of Flux as the place where your repetitive work goes to become automatic. You build a workflow once, and from then on it runs itself — either on a schedule (every Friday at five, the first of the month, every morning) or whenever something happens (a new client is added, a deal goes quiet, an invoice is marked paid).
There are two ways to set one up, and you can use whichever feels natural. You can type what you want in plain English — "Every Monday morning, pull last week's sales totals and email me a short summary" — and let it draft the workflow for you. Or you can lay it out visually on a simple canvas, dragging the steps into the order you want, like sketching the process on a whiteboard. No code. No cryptic settings. Just a clear picture of when this happens, do that.
The quiet magic is that Flux isn't off in a corner by itself. Because it lives on the same platform as the rest of your work, it can actually reach the things you care about. The report can read your real numbers. The welcome message can pull the new client's real name. The follow-up can look at the actual deal that's gone cold. You set it up once with the information already in the system, so nothing has to be re-typed or copied across — the thing you used to do by hand on Friday now happens without anyone touching it.
Picture your old Friday. Now picture it gone. The report lands in your inbox at 5:01, already formatted, already summarized, while you're putting your coat on. The new client gets their welcome note the moment they're added, warm and on time, every single time, whether you're at your desk or on a beach. The deals that drift get a gentle nudge before they go cold, instead of after. You didn't do any of it. You just decided, once, that it should happen.
But will it act without me?
Here's the worry that probably surfaced the second I said "runs itself," and it's the right worry to have. The whole appeal of automation is that it acts on its own — which is also exactly what makes it nerve-wracking. You don't want to wake up to discover that a workflow emailed the wrong number to a hundred customers, or sent a payment reminder to your best client the day after they paid.
So SMAIVIZ is built around a simple principle: automation handles the routine, but a human stays in control of anything that matters. When a workflow is about to do something sensitive — send a message to a customer, move money, change something that's hard to undo — it can pause and wait for your approval. You get a clear summary of what it's about to do, and nothing happens until you say go. Routine, invisible steps flow on their own. Consequential ones wait for a person. You decide which is which.
And every workflow keeps a clear record of what it did and when — so you're never guessing whether the welcome emails went out or who approved the thing that went out. If something looks off, you can trace it back. That's the difference between automation you have to babysit and automation you can actually trust: you're not handing over the keys, you're handing over the chores, and keeping the keys.
One more thing worth knowing: you can bring your own AI model. The intelligence behind your workflows isn't a black box you're locked into — it works with the model you choose.
Questions you're probably asking
Do I need to know how to code?
No. You can describe what you want in plain English and let SMAIVIZ draft the workflow, or arrange the steps visually on a canvas. Most people start by typing a sentence and adjusting from there. There's nothing technical to learn before you get value.
What can actually trigger a workflow?
Two kinds of things. A schedule — every Friday, the first of the month, every morning — or an event, like a new client being added, a deal going quiet, or an invoice being marked paid. You set the "when," and the workflow takes care of the rest.
I'm worried it'll send something wrong to a customer.
That's exactly why sensitive steps pause for a human approval. Before a workflow messages a customer or does anything hard to undo, it can stop and show you a summary, and wait until you approve. The routine steps run on their own; the consequential ones wait for you.
Will it work with the tools and data I already use?
Because automation lives on the same platform as your work, your tasks, customers, and finances, it can use the information that's already there — no copying between tools, no re-typing. You build a workflow once and it draws on your real, current data each time it runs.
How do I know what it actually did?
Every run leaves a clear record — what happened, when, and what was approved. If you ever need to check whether something went out or trace a result back, it's there. You're never left guessing.
You don't have to automate everything at once. Pick the one task you dread most — the Friday report, the welcome email, the follow-up you always forget — and let that one run itself. Then notice how it feels to close your laptop on Friday and know the report is already handled. That feeling is the whole point. Start free at smaiviz.com.
It's Sunday evening. The week isn't over and the next one hasn't started, but already there's a low hum in the back of your mind — the email you meant to send, the thing you promised someone on Tuesday, the report that's either nearly done or barely begun, you genuinely can't remember which. You're not working. You're just carrying it. And the more you try to hold it all in your head so nothing slips, the heavier your head gets.
The same work, moved out of your head and your chat threads into one shared, ordered place.
Most of us don't have a productivity problem. We have a storage problem. The work itself is fine — you know roughly what needs doing. What's exhausting is that it lives everywhere at once: a sticky note on the monitor, a starred message you'll never find again, a half-finished list in your notes app, and the rest of it nowhere at all except the anxious part of your brain that refuses to clock off. No single place shows you the whole picture, so you become the place. You're the database. And a human being is a terrible database.
The quiet cost of holding it all in your head
When work is scattered, the damage isn't just the occasional dropped ball, though those hurt — the missed follow-up, the deadline that arrived a day early. The deeper cost is subtler. Priorities blur, because you can't compare two things you can't see side by side, so you end up doing whatever shouted loudest most recently. You repeat yourself, asking someone what they're working on because there's no other way to know. And you never fully rest, because part of you is always on guard duty, making sure nothing important falls through a crack you can't quite see.
If you lead a team, multiply all of that. Now you're not just holding your own work — you're holding everyone's, half-guessing who's blocked, who's idle, what's actually moving. Status updates become a tax everyone pays in meetings nobody enjoys. The honest question underneath it all is simple: what if you didn't have to be the one remembering everything?
Get the whole picture into one place first
The relief, when it comes, doesn't come from a clever trick. It comes from finally seeing everything in one frame. This is where Kriya earns its keep. You take the goal you're carrying — ship the launch, close out the quarter, onboard the new client — and break it into a project with real tasks underneath it. Each task gets the two things that turn a wish into work: an owner and a due date. Suddenly the fog has edges.
What changes first is not your output but your shoulders. The vague dread shrinks because the unknown shrinks. You can see, in one view, what's yours, what's someone else's, what's due, and what's quietly slipping. Nothing is hiding in a thread anymore. When you want to push hard on one thing, you pull a slice of that work into a focused sprint — a short, bounded run where the team agrees on what "done" looks like and protects the time to get there — instead of nibbling at forty things and finishing none.
Turn a goal into a project, and the project into tasks anyone can see.
Give every task an owner and a due date, so "who's got this?" stops being a question.
Run focused sprints when something matters more than everything else this week.
Now hand some of it off — to a person or an AI agent
Here's the part that genuinely changes the math. In Kriya, a task isn't only something a human can own. Any task can be assigned to a teammate — or to an AI agent that can actually do the thing. The first draft of the summary, the data pulled together into a tidy table, the routine write-up you do the same way every week: those can go to an agent the same way you'd hand them to a colleague, with the same owner-and-due-date clarity. You stay the one deciding what gets done and why. The agent picks up the doing.
The feeling that creates is hard to describe until you've had it: your plan and your team move together, and your team now includes hands that don't get tired. You're not managing two systems — one for people, one for AI. It's one list. Some rows are people, some are agents, all of them are moving. The work that used to sit on your own plate because "it's faster to just do it myself" finally has somewhere else to go.
Handing off without losing the thread
A reasonable worry shows up right about here: if I let an agent loose on my work, do I lose control of it? It's the right question to ask, and the answer is the part that should let you exhale. Anything sensitive waits for a human to approve it before it happens — nothing consequential slips out the door on its own. And every step leaves a clear trail: what was done, by whom or by which agent, and when. You can hand work off without handing off your judgement. The plan stays yours; the busywork stops being yours alone.
And because Kriya doesn't live on an island, the same project sits next to the rest of your business — your customers, your people, your numbers — under one login with shared data. The task to follow up with a client isn't divorced from the client; it's part of the same connected picture, with the same human approval and the same audit trail running underneath all of it.
Questions you're probably asking
I've tried a dozen to-do apps and abandoned them all. Why would this stick?
Most lists die because they're one more place to check that gives nothing back. Kriya is different in one specific way: it doesn't just hold your tasks, it can do some of them. When a tool actively takes work off your plate — by routing the right tasks to a teammate or an AI agent — you keep coming back, because it's earning its place rather than just storing your guilt.
What kind of work can I actually hand to an AI agent?
The repeatable, structured pieces: drafting a summary or a first version, gathering scattered information into one tidy place, the recurring write-up you do the same way each time. You assign it like you'd assign any task, with an owner and a due date — the agent just happens to be the owner. You review what comes back before anything sensitive goes anywhere.
Will I lose control of work once an agent is involved?
No. Anything sensitive pauses for a human to approve it, and every action leaves a clear record of what happened and who or what did it. You're always the one deciding what gets done; the agent handles the doing within the limits you set.
Does this only work for big teams?
It works just as well for one person. A solo doer gets the same relief — everything in one place, focused sprints, and the option to offload routine tasks to an agent. The picture simply grows with you as your team does, people and agents on the same list.
So picture the next Sunday evening, the one where your work isn't riding around in your head anymore — it's sitting in one place with owners and due dates, and a few of those owners are agents quietly working through the routine parts while you actually rest. That's not a fantasy of a perfectly organised person you'll never be. It's just what happens when the work has somewhere to live that isn't you. Start free at smaiviz.com.
It's a Tuesday, and a name you recognize is calling. You let it ring once while you scramble. Who is this, again — the one who wanted the larger order, or the one waiting on a revised quote? You open your inbox and search their name. Forty messages come back, half of them from a colleague who left in the spring. You open a spreadsheet. The last note says "follow up next week," dated a month ago, in a column nobody has touched since. You pick up anyway, warm and confident, and ask the question you swore you'd never ask again: "So, remind me where we left off?"
The context isn't missing. It's scattered. The fix is gathering it into one place.
Here's the part that stings: you didn't drop the ball because you stopped caring. You dropped it because caring required reassembling a story from six places every single time, and there are only so many hours in a day. The promise you made on a call lived in your head. The colleague's reply lived in their sent folder. The renewal date lived in a spreadsheet that someone forked into a second version, and now no one is sure which is current. None of those pieces is wrong. They're just apart.
The quiet cost of scattered context
A relationship that lives in fragments is a relationship that leaks. Not loudly — there's rarely a single dramatic failure you can point to. It's a slow drip. The follow-up that nobody owned because everyone assumed someone else had it. The deal that went cold not from a "no" but from silence, because the next step was written on a sticky note that got covered by another sticky note. The customer who told you, gently, that they'd "actually mentioned this before," and they had, and you'd just never seen it because it was in a thread you weren't on.
What it costs you isn't only deals. It's the way you feel walking into every conversation slightly behind, bracing for the thing you forgot. It's the founder who can't take a holiday because the pipeline only really exists in their memory. It's the new hire who takes three months to be useful because the history they need isn't anywhere they can read it. Scattered context taxes your confidence, and confidence is most of what selling actually is.
So here's the question worth sitting with: what would change if, the moment any customer's name came up, you already knew the whole story — every conversation, every promise, every open thread — without hunting for it?
One place that holds the whole story
This is the gap SMAIVIZ Orbit is built to close. Orbit is a place where every company and every contact lives as a single record with its full history attached — the emails, the calls, the notes, the meetings, who said what and when. Open a customer and the relationship is just there, in order, ready to be read. No reconstructing. No "let me dig around." The story stays with the person it belongs to.
Your pipeline gets the same treatment. Instead of guessing which deals are alive from a spreadsheet you don't quite trust, you see every deal on one board, sitting in the stage it's actually in. You can tell at a glance what's moving, what's stalled, and what needs you today. The board is the truth, and everyone looking at it sees the same truth.
And when a new lead arrives, you don't retype anything. You turn that lead into a full customer record in a single click — the contact, the company, the early history, all carried over intact. The moment of interest becomes a relationship you can actually steward, instead of a name you'll mean to follow up with and won't.
People tend to describe the difference in plain terms:
"I stopped asking people to remind me where we left off."
"Anyone on the team can pick up a conversation cold and sound prepared."
"Nothing falls through the cracks anymore, because there are no cracks — it's one place."
Because Orbit shares a single login and the same data with the rest of SMAIVIZ, the context doesn't stop at the edge of your CRM. The task you create to follow up shows up in your work. The deal that closes can flow into your books. Your customers are understood the same way across everything you run, instead of being re-explained from scratch in every tool.
Quietly automatic, never out of your hands
Some of the upkeep can run on its own. You can have routine follow-ups drafted, records kept tidy, and the busywork around a deal handled without you typing it out each time — described in plain English or laid out on a simple visual canvas. The goal isn't to take you out of the loop. It's to take the dread out of the parts that never needed your judgment.
For anything that actually touches a customer — a message going out, a meaningful change to a deal — a human stays in control. Sensitive steps wait for your approval before they happen, so nothing reaches a customer that you didn't choose to send. And every action leaves a clear trail: who did what, and when. If you ever need to retrace how a deal moved, the record is right there. You can bring your own AI model into the mix, too, so the help you get works the way you want it to.
Questions you're probably asking
I already have my customers in spreadsheets. Do I have to start over?
No. Orbit is designed to gather what's scattered, not make you rebuild it. You bring your existing companies and contacts in, and from then on they live in one place with their history attached — so the spreadsheet stops being the thing you depend on and quietly becomes a backup you don't need.
What happens to a lead once it comes in?
In Orbit, a lead is one click away from becoming a full customer record — the contact, the company, and the early context carried over. The moment of interest turns into a relationship you can actually follow up on, instead of a name you mean to get back to.
Will the rest of my team see the same picture I do?
Yes. The pipeline board and each customer's history are shared, so anyone can open a record and pick up the conversation prepared. With one login across SMAIVIZ, that same context carries into your tasks and your books as well.
If automation is involved, am I still in control of what reaches a customer?
Always. Sensitive steps — like a message going out — wait for your approval before anything happens, and every action is recorded in a clear audit trail. The automation handles the busywork; the judgment stays with you.
Imagine the next time that familiar name lights up your phone. You answer on the first ring, the whole story already in front of you, and you pick up exactly where you left off — because it never went anywhere. That's the quiet relief Orbit is for: not more software to manage, but fewer things to remember and worry about. Start free at smaiviz.com.
Your marketing lives in a dozen places at once. The blog draft is in one tool, the social posts in another, the email in a third, the ad campaigns in a fourth, and the spreadsheet that's supposed to tell you what actually worked is somewhere in your downloads folder, three versions out of date. You're not short of ideas. You're short of a single place to run them from — and the hours it takes to stitch it all together by hand. What if you could describe the campaign you want, and a team of specialists drafted the whole thing for you, ready to review?
Content, social, email, ads and web — planned from one goal, with a checkpoint before anything is published or paid for.
Here's what marketing actually feels like for most small teams: it's less a strategy and more a juggling act. You're the strategist, the copywriter, the designer, the person who remembers to actually post the thing, and the analyst who's supposed to figure out whether any of it worked. Each of those is a real job. You're doing all of them in the cracks between everything else. So the blog goes quiet for a month. The newsletter goes out late, or not at all. The ad keeps spending on the audience you meant to switch off two weeks ago.
None of that is a failure of effort. It's a failure of coordination. The work is scattered across tools that don't talk to each other, so nothing builds on anything else. You write a great blog post and then re-write it from scratch as a social thread, and again as an email, because there's no thread connecting them. The result is a lot of motion and not much momentum.
What if marketing had a cockpit?
This is the idea behind Echo, the marketing part of SMAIVIZ. Instead of a drawer full of disconnected tools, you get one place to plan and run everything — a cockpit. A shared calendar shows every post, email, ad and content piece on one timeline. A campaign groups all the pieces that belong together under a single goal and budget, so the blog post, the social series, the nurture emails and the paid ads are all working toward the same thing instead of drifting off in separate directions.
And you're not doing it alone. Echo gives you a marketing copilot and a small team of AI specialists — one that's good at content and SEO, one for social, one for email and lifecycle, one for ads, and an analyst that watches the numbers. You describe the goal in plain language — "launch our new pricing tier to existing customers over the next two weeks" — and the copilot turns it into a plan, then hands each piece to the right specialist to draft. You review on-brand drafts instead of starting from a blank page.
On-brand, because it knows your brand
The reason the drafts don't sound like generic robot copy is that the agents are grounded in your brand — your voice, your guidelines, the way you talk about what you do. Set that up once and every draft, in every channel, comes back sounding like you. The blog post and the social version and the email all share a thread, because they came from the same plan and the same voice, not from three different late nights.
Because it all lives on one platform, the pieces actually connect. A campaign in marketing can tie back to the real customers in your CRM. The analyst can join your ad spend to the deals that actually closed, so "is this working?" gets a real answer instead of a shrug. The content you publish, the forms people fill in, the links they click — they feed back into one clear picture of which channels are pulling their weight.
But it won't post or spend without me
Here's the worry the moment you hear "AI agents run your marketing": you do not want to wake up to find a half-finished draft posted to your public feed, or an ad campaign that quietly burned the budget overnight. The whole point of marketing is that it's seen — which is exactly what makes handing it to automation nerve-wracking.
So Echo draws a hard line. The agents plan and draft freely — that part is fast and autonomous, because a draft can't hurt anyone. But anything that sends an email, publishes a post, or spends ad money stops and waits for your approval. You see exactly what's about to go out, and nothing leaves the building until you say go. You're not handing over your brand. You're handing over the busywork of producing the work, and keeping the final yes for yourself.
One more thing worth knowing: you can bring your own AI model. The intelligence behind your marketing isn't a black box you're locked into — it works with the model you choose.
Questions you're probably asking
Do I have to use every channel?
No. Start with the one or two that matter most — maybe content and email — and add social or ads when you're ready. The calendar and campaigns work the same whether you run one channel or all of them.
Will the AI post things on its own?
Drafting and planning happen automatically, but anything that publishes, sends, or spends pauses for your approval first. You review a clear summary and nothing goes out until you approve it.
Will the copy actually sound like us?
You set your brand voice and guidelines once, and every draft is grounded in them — so the blog, the posts and the emails share one consistent voice instead of reading like generic AI text.
How do I know what's working?
A built-in analyst ties your activity and spend back to real results — which channels and campaigns drove conversions — so you can put more behind what works and stop what doesn't.
Does it connect to the rest of my business?
Yes. Because marketing lives on the same platform as your customers, work and finances, campaigns can reference real customer records and join spend to closed deals — no copying between tools.
You don't have to run all your marketing through one place on day one. Pick the campaign you've been putting off — the launch, the newsletter you keep meaning to restart — describe the goal, and let the agents draft it while you keep the final say. Then notice how it feels to review a finished plan instead of staring at a blank page. Start free at smaiviz.com.
It is Tuesday morning and you are scrolling through the export one more time before the campaign goes out. There she is again: Jennifer Walsh. And right below her, Jen Walsh. And a few rows down, J. Walsh, with a work email that bounced last spring. You are almost certain all three are the same person. You are not certain enough to merge them, and you are not certain enough to leave them. So you do what you always do. You hover, you sigh, and you hit send anyway, hoping she only gets the email once.
Four scattered records about the same person, stitched into one profile you can actually trust.
None of this is your fault. The records came from everywhere: the signup form, the trade-show scanner, the support inbox, the spreadsheet a colleague kept on their own machine for two years, the checkout system that never asked for a last name. Each one captured a sliver of a real human being, and none of them talked to each other. So now you have twelve rows where there should be one person, and no honest way to tell which version is the truth.
The quiet cost of a list you can't trust
On paper it looks like a tidy-up task, something for a slow afternoon. In practice it bleeds into everything. You email the same person three times because they sit in three segments. You greet someone as "Hi {FirstName}" because the merge field came up empty. You send a winback offer to a loyal customer and a loyalty perk to someone who churned in February. Each one is small. Each one chips at the thing you can least afford to lose, which is the sense that you actually know the people you are talking to.
And it costs you more privately than that. When the numbers come up in the meeting and someone asks how many active customers you really have, you give a range, not a figure, because you know a chunk of the list is duplicates and ghosts. You stop trusting your own dashboard. You start padding every estimate. Slowly, quietly, you become a person who does not believe their own data, which is an exhausting way to do good work.
So here is the question worth sitting with: if you could look at any customer and see one clear, complete, current picture of them, how much of the second-guessing would simply fall away?
What "one clear customer" actually means
This is the problem Prism, the customer-data part of SMAIVIZ, was built to take off your shoulders. The idea is simple to say and surprisingly hard to do by hand: take every scattered record about the same person, from every source you have, and resolve them into a single golden profile.
Prism reads the signals a careful human would — the same email turning up twice, a phone number shared across rows, a name spelled four ways — and decides, with care, that Jennifer and Jen and J. Walsh are one person. It stitches them into one profile. Where pieces are missing, it can enrich the record with the detail that was scattered across your other sources, so you are not left staring at half a customer. And it keeps that profile clean over time, instead of letting the mess quietly creep back the moment new data lands.
What you are left with is the thing you have wanted all along:
One profile per real person, not five rows wearing the same face.
The current email and number surfaced, with the dead ones set aside.
A count you can say out loud in a meeting without flinching.
From a clean profile to a campaign you can stand behind
A tidy list is only worth something if it changes what you do next. Once your profiles are clear, you can build an audience from what is actually true — the customers who bought in the last quarter, the ones who went quiet, the segment you have been meaning to reach but never trusted enough to target. Then you can activate that audience, send it where the work happens, and feed it straight into Orbit, the CRM, so your sales and success people are working from the same clean picture you are.
Because Prism sits alongside the rest of SMAIVIZ under one login and shares its data, there is no more exporting, reconciling, and re-importing across tools that each believe something different. The profile you trust in Prism is the profile your colleague sees in the CRM. One source of truth, not twelve copies drifting apart.
Cleaning data without losing the plot
There is a fair worry hiding under all of this. Merging customer records and reaching out to audiences is sensitive work, and software that does it carelessly can do real damage — wrong merges, a message to someone who asked to be left alone. So SMAIVIZ is built to keep you in the chair.
On anything sensitive — an audience push, a bulk action, a step that touches real people — a human approval stands between the suggestion and the act. Prism can propose that two records are the same and that a segment is ready to go, but you decide. And everything that happens leaves a clear, readable audit trail, so when someone asks why a customer was merged or where an audience came from, the answer is right there, not a shrug. Your data is handled responsibly, and you can bring your own model, so the intelligence works on your terms.
When was the last time you felt fully calm sending a campaign, knowing the list underneath it was actually clean?
Questions you're probably asking
Will it merge records I didn't want merged?
No. Prism proposes matches based on the signals it reads, but on sensitive actions a human approval stands in the way, so you confirm before anything is combined. And every merge is recorded in a clear audit trail, so it can be reviewed or traced later.
What about the records that are just missing information?
When pieces of a profile are scattered across your other sources, Prism can enrich the record by pulling that detail together, so you end up with one current, complete picture instead of half a customer.
Does the clean profile reach my sales team, or just marketing?
Both. Prism and Orbit, the SMAIVIZ CRM, share data under one login, so the golden profile you trust is the same one your sales and success colleagues see. No exporting and re-importing between tools.
Can I build audiences and send them somewhere?
Yes. Once profiles are clean you can build an audience from what is true, then activate it and feed it into your CRM, with a human approval on the sensitive steps and a record of what was done.
Is my customer data handled responsibly?
It is the priority. Sensitive actions require human approval, everything leaves an audit trail, and you can bring your own model so the intelligence works on your terms.
You do not need a perfect list. You need one you can believe, where Jennifer is one person and the count is honest and the campaign goes to the people you meant. That is the quiet relief on the other side of this: not more data, but data you finally trust. Start free at smaiviz.com.
It's a Tuesday, and you're reading a customer reply that someone on your team already sent. It's good — friendly, on-brand, exactly what you'd have written. The trouble is, you don't recognize it. You scroll up. No name you know. After a minute it lands: a person didn't send this. One of the AI agents did. You're glad it went out. You're also, quietly, a little unsettled — because if someone asked you right now who is responsible for that message, you'd have to admit you don't actually know.
Solid boxes are people, dashed boxes are AI agents — and every agent reports to a human who owns it.
For most of us, AI didn't arrive as a grand decision. It arrived one helpful task at a time. Someone wired up a bot to draft replies. Someone else let one chase overdue invoices. A third quietly drafts the weekly summary nobody used to have time for. Each of these was a small, sensible win. Added together, they've changed something bigger: a real share of the work in your business is now done by workers who don't show up in any directory, don't have a manager, and never appear in the room when you ask "who's handling this?"
The team you can only half-see
The discomfort isn't that the AI is bad at the work. Often it's good — that's what makes this strange. The discomfort is that your team has become half-visible. You can name your people, see who reports to whom, and sense when someone's drowning. The AI side is a fog. You're not sure how many agents are running, what each one is permitted to touch, or who would notice if one started behaving oddly. When a customer asks why they got a particular message, the honest answer is a shrug and a search.
That fog has a quiet cost, and it isn't mostly about the AI. It's about you. Accountability is the thing that lets a manager sleep. When every piece of work traces back to a person who owns it, you can delegate without dread. When some of the work traces back to nobody, you carry a low background hum of responsibility for things you can't see — which is exhausting in a way that's hard to name and harder to fix by working harder.
So here's the turning-point question: what if your AI agents were simply members of the team — listed, owned, onboarded, and accountable, exactly like the people next to them? Not magic. Not hidden helpers. Just teammates you can point to.
Put everyone in one directory
This is the idea behind Hive, the part of SMAIVIZ for managing your team — and in Hive, "team" means people and AI agents in one place. Open the directory and you see everyone: Maya in support, Dev in sales, and right alongside them the Reply Agent and the Invoice Agent, each on its own card. Same org chart. Same sense of structure. The difference is no longer "human versus mystery" — it's just one team you can read top to bottom.
Crucially, every AI agent has an owner. The Reply Agent reports to Maya; the Invoice Agent reports to Ops. That single fact dissolves most of the unease. The next time a message goes out that you don't recognize, you don't shrug — you look at the card, see who owns it, and you have a person to talk to. The agent did the work; a human stands behind it. Responsibility stops floating.
Onboard an agent like you'd onboard a hire
You wouldn't hand a new employee the keys to everything on day one. You'd say: here's your role, here's what you're responsible for, here's what's not yours to touch. AI agents deserve the same care, and treating them casually is exactly how they drift into doing things nobody intended.
In Hive you onboard an agent the way you'd onboard a person. You give it a role and a clear remit. You decide what it can see and do. You write down, in plain language, what "good" looks like for it. From then on its scope isn't tribal knowledge living in one engineer's head — it's there for anyone to read. And because everything in SMAIVIZ sits behind one login and shares the same data, an agent you onboard in Hive is the same teammate that shows up doing work over in your CRM or your books. You define it once; it's legible everywhere.
A role and an owner, so accountability is never in question.
A defined remit, so "what's it allowed to do?" has a written answer.
A shared standard, so a quick check-in can tell you whether it's still pulling its weight.
See who's actually overloaded
Capacity planning quietly breaks when half the team is invisible. You look at your people, see they're stretched, and assume the team is at its limit — while three agents hum along with room to spare, or one is silently buried under more than it can handle well. You can't plan around what you can't see.
When people and agents share one directory, capacity becomes a real picture again. You can ask the honest questions: Is anyone overloaded? Where's the slack? Should this new work go to a person or to an agent — and which one? Hive also keeps room for lightweight performance check-ins, for the AI as much as the humans. Not a heavy review cycle — just a regular, simple "is this teammate still doing what we need?" applied evenly across the whole team. Have you ever actually checked whether your busiest AI agent is doing good work, or only assumed it because nothing broke?
Stay in control of the parts that matter
Visibility on its own would still leave you nervous about the consequential moments — an agent issuing a refund, sending something to your whole customer list, changing a number that matters. So SMAIVIZ keeps a human at those gates by design. Sensitive actions can pause for a person's approval before they happen, rather than after. The agent does the heavy lifting; you keep the final yes.
And everything an agent does leaves a clear trail. Who did what, when, on whose behalf — written down and reviewable. That trail is what turns "an AI sent this" from an unsettling sentence into an ordinary one. You don't have to choose between letting AI help and knowing what it did. You get both, and that's the whole point.
Questions you're probably asking
Do I have to be technical to set up an AI agent here?
No. In Hive you give an agent a role, an owner, and a remit in plain language — the same things you'd tell a new hire. SMAIVIZ handles the rest, and you stay in the manager's seat, not the engineer's.
What stops an AI agent from doing something it shouldn't?
Two things. You define each agent's remit when you onboard it in Hive, so its scope is explicit. And for sensitive actions, SMAIVIZ pauses for a human's approval before anything happens — so the consequential moments always pass through a person.
We already use AI tools all over the place. Can I bring them into one view?
That's exactly the problem Hive is built for. Instead of agents scattered across disconnected tools, you put people and agents in one shared directory and org chart, each with an owner, so the whole team is finally visible in one place.
If something goes wrong, can I tell what an agent actually did?
Yes. Every AI action in SMAIVIZ leaves a clear, reviewable trail — who did what, when, and on whose behalf. You can always trace an action back to the agent that took it and the human who owns it.
Can I use my own AI model?
You can bring your own model to SMAIVIZ. The agents work the same way in your team's directory regardless of which model is behind them — the structure, ownership, and accountability stay consistent.
Your team didn't ask to become half-visible; it happened one helpful task at a time. The fix isn't to pull the AI back out — it's to bring it the rest of the way in, until every teammate, human or not, has a name, an owner, and a place on the chart. When that's true, the Tuesday-morning message you don't recognize stops being a small worry and becomes just another piece of work you can trace, in seconds, back to a person who stands behind it. Start free at smaiviz.com.
It's the last Sunday of the month, and there's a drawer you've been avoiding. Inside it: a crumpled fistful of receipts, a bank statement you half-skimmed, and a sticky note that says "ask about the November thing." You make a cup of tea you don't really want, open a spreadsheet, and feel the familiar tightening in your chest. Somewhere in all of this is the answer to a simple question — did the business make money this month? — and you genuinely don't know.
The same numbers — just somewhere you can finally read them.
You are not bad with money. You started this business because you were good at something — the craft, the product, the service, the relationships. Bookkeeping was never the dream. But somewhere along the way it became this quiet, accumulating debt of attention, and now month-end arrives like a small monthly reckoning you'd rather not have.
The cost of not quite knowing
The dread isn't really about the receipts. It's about what they represent: a fog over your own business. When you can't see clearly, you make decisions slowly and nervously. Can you afford to hire? You're not sure. Should you take that big project that needs cash up front? You hesitate. Is that client who "always pays eventually" actually costing you? You suspect so, but you can't prove it.
There's a particular kind of shame that comes with this — the feeling that a "real" business owner would have this handled. So you keep the drawer closed a little longer. You tell yourself you'll sort it out properly next quarter. And the numbers drift further from anything you could explain to an accountant, a lender, or yourself.
The cost is rarely a single dramatic mistake. It's the slow tax of operating half-blind: the late-paid invoice you forgot to chase, the duplicate subscription nobody caught, the quarter you thought was fine that wasn't. None of it screams. It just quietly takes.
So here's the turning-point question worth sitting with: what would change if, on any given Tuesday, you could open one screen and actually trust the number you saw there?
What "proper books" actually means
Proper books sound intimidating, like something only a trained accountant can keep. But the bones of it are simpler than the word "accounting" suggests. You need a place where every pound or dollar has a home — a chart of accounts — so money in and money out land in sensible buckets instead of a shoebox. You need to send invoices and know which ones are still owed to you. You need to track the bills you owe, so nothing sneaks up. And then you need reports that turn all of that into a plain answer: here's what you earned, here's what you spent, here's where you stand.
This is the part that's been missing — not the data, but a place that organizes it for you and shows it back in a language you can read. That's the gap SMAIVIZ built Artha to close.
Artha is the finance and accounting part of SMAIVIZ. It keeps a real chart of accounts in the background, lets you send invoices and track the bills you owe, and turns the whole picture into financial reports that don't require a degree to understand. Instead of you wrestling numbers into a spreadsheet at 9pm, the structure is already there. You're reading the story your business is telling, not assembling it from fragments.
From a drawer of receipts to a Tuesday you can read
Picture the same month-end, but different. The invoices you sent are recorded the moment they go out, and you can see at a glance which are paid and which are still outstanding. The bills you owe are listed with their due dates, so the surprises stop being surprises. And when you want to know how the month actually went, you open a report and read it — revenue, expenses, what's left — in plain rows that add up.
What changes most isn't the bookkeeping. It's you. The drawer loses its power. The Sunday-night tightening eases, because the question "are we okay?" finally has somewhere to be answered. People describe it less as "I love accounting now" and more as "I stopped being afraid to look."
You can see who owes you and how much, without reconstructing it from memory.
You know which bills are coming, before they become a scramble.
You can hand a clean, readable report to your accountant — or a lender — without apologizing first.
And because SMAIVIZ is one platform with shared data behind a single login, the money doesn't live on an island. The customer you invoice in Artha is the same customer your team already knows in the rest of SMAIVIZ. Bring your own model if you like; the books stay yours.
Stay in control of anything that touches money
Here's the part that matters most when software gets close to your finances: nothing money-related should run on autopilot just because it can. In SMAIVIZ, anything that touches money waits for a human approval before it happens. If a payment-related action is about to run, or an automated reminder is set to go out to a customer who's overdue, it pauses for a person to say yes.
And every one of those approvals leaves a clear record — who approved what, and when. That audit trail isn't bureaucracy; it's peace of mind. Months later, when you or your accountant asks "why did this happen?", the answer is right there. You get the help of automation without ever handing over the keys.
Ask yourself honestly: how much of your month-end dread is really just the fear of something happening to your money that you didn't see and can't explain? Take that fear off the table, and the whole task gets lighter.
Questions you're probably asking
I'm not an accountant. Will I understand any of this?
That's exactly who Artha is built for. The chart of accounts works quietly in the background — what you see are invoices, bills, and reports written in plain language. You read where you stand; you don't have to assemble it.
Will it do things to my money without asking me?
No. In SMAIVIZ, anything money-related waits for a human approval before it runs, and every approval is recorded — who said yes, and when. Automation helps you; it never acts on your finances on its own.
Can it chase the invoices people forget to pay?
Yes. Flux, the automation part of SMAIVIZ, can send overdue-invoice reminders for you — and because that touches a customer and their money, it pauses for your approval before going out.
Does my finance data connect to the rest of my business?
It does. SMAIVIZ is one platform with one login and shared data, so the customers you invoice are the same records your team works with everywhere else — no copying between disconnected tools.
Do I have to switch to a different AI model?
No. SMAIVIZ lets you bring your own model, and your books and records stay yours throughout.
You don't have to become a finance person. You just deserve to stop dreading the drawer — to open one screen, read a clear number, and know where your business stands. That's a quieter, steadier way to run things, and it's within reach this month. Start free at smaiviz.com.
It's a Tuesday afternoon and you open a chat with an AI tool, the way you have a hundred times now. You ask it to pull together the highlights from three long reports, draft a follow-up note, and figure out who still owes you a reply. A few seconds later, back comes a paragraph that's genuinely good. Clear. Well-organized. You nod. And then you sit there, because the answer is just words on a screen, and the actual work — the opening of the documents, the writing into the right place, the chasing of the right people — is still entirely, stubbornly yours.
The gap most people feel with AI: a great answer is not the same as the work being done.
If you've felt that particular flatness, you're not imagining it. The current crop of AI tools are wonderful conversationalists. They explain, they summarize, they suggest. But they hand everything back across the counter and let you carry it home. It's a little like hiring a brilliant intern who reads every document, has an opinion on everything, drafts beautifully — and then never once files anything, sends anything, or remembers what you asked yesterday. You get the thinking. You keep the doing.
And the doing is where the day actually goes. Not the question "what should this email say," but the forty small follow-throughs: opening the thread, pasting the draft, fixing the name, checking it against the notes, sending it, marking the task done, moving to the next one. Each is trivial. Together they are your afternoon. The quiet cost of clever-but-passive AI is that it relocates the interesting part — the answer — and leaves you with all the tedious parts that surround it. Over a week, that adds up to a strange exhaustion: you had help, and you're still tired.
What changes when help can follow through
So here's the question worth sitting with: what if the help didn't stop at the answer? What if you could hand off not just "tell me what to write" but "write it, put it where it belongs, and tell me when it's ready for me to look at"?
That's the shift this is about. In SMAIVIZ, you don't just talk to an AI — you build an assistant of your own. You give it a job: research a topic, draft the recurring report, summarize the long thread, answer the routine question that lands in your inbox every Monday. Then you decide where it works. You can keep it in a chat, the familiar back-and-forth, asking and refining as you go. Or you can put it on a schedule, so the thing you keep meaning to do quietly gets started without you having to remember it at all.
The part that matters most is that your assistant can pick up real tasks, not just describe them. When work is tracked as actual tasks, your assistant can take one on — research it, draft the output, move it forward — and report back. It stops being a window you visit and becomes a colleague who's already started.
You build it; it's yours
An assistant you create is shaped by you. You tell it what it's for and how you like things done — the tone of your writing, what "good" looks like, what to leave alone. Over time it fits the contours of your work rather than the generic average of everyone's.
You also choose the mind behind it. SMAIVIZ lets you bring your own AI model, so you're not locked into one provider's idea of how an assistant should think. Prefer one model for careful drafting and another for quick summaries? That's your call. The assistant is the helper; the model is the engine, and you pick the engine.
A few of the everyday jobs people hand to one:
Reading three long documents and coming back with the five things that actually changed.
Drafting the weekly update so you're editing instead of staring at a blank page.
Answering the same routine question that arrives over and over, in your words.
Quietly preparing the recurring report on schedule, so it's waiting when you sit down.
None of these are dramatic. That's the point. They're the small, repeated things that wear a groove in your week, and they're exactly the things worth handing off.
You stay in control of the doing
There's an obvious worry here, and it's the right one to have: if an assistant can actually do things, what stops it from doing the wrong thing? The answer is that you're never handing over the keys. Anything sensitive — sending the message, changing the record, acting on the world rather than just thinking about it — waits for your approval. The assistant prepares; you say yes. You see what it's about to do before it does it, and nothing irreversible happens behind your back.
And because everything runs in one place under a single login, with your data shared across the work rather than scattered, you get a clear trail of what was done, when, and by whom — assistant or human. That's not bureaucracy. It's the thing that lets you trust delegation: you can always look back and see exactly what happened. Help you can't audit isn't help you can lean on.
Ask yourself honestly: how much of your week is answers you already have, waiting on someone to carry them the last few steps? And what would it feel like to close the laptop knowing those steps were already underway?
Questions you're probably asking
How is this different from the AI chatbot I already use?
A chatbot answers and stops. A SMAIVIZ assistant answers and then follows through — it can pick up a tracked task, draft the output, and move it forward, reporting back when it's ready for you. You build it once for a specific job rather than re-explaining yourself every session.
Do I have to be technical to build one?
No. You describe what the assistant is for and how you like things done in plain language, then put it to work in a chat or on a schedule. SMAIVIZ handles the rest; you stay the editor and the approver.
Can I use my own AI model?
Yes. SMAIVIZ lets you bring your own model, so you choose the engine behind your assistant — and you can match different models to different kinds of work.
What keeps it from doing something I didn't want?
Anything sensitive waits for your approval before it happens, and SMAIVIZ keeps a clear record of every action. The assistant prepares the work; you decide what actually goes out.
Does it work with the rest of my tools?
Your assistant lives in the same SMAIVIZ workspace as your tasks, customers, and books, under one login with shared data. So when it picks up a task or pulls together a summary, it's drawing on the same information you are.
You don't need an AI that's a better talker. You need one that closes the distance between the answer and the work — and lets you keep your hand on the wheel while it does. Build an assistant that's actually yours, point it at the things that wear you down, and let it start before you do. Start free at smaiviz.com.
It's a little after nine, and you haven't done any real work yet. You've opened the CRM to check whether that deal moved. You've flipped to the tasks app to see what your one part-time helper is meant to be doing today. You've got a spreadsheet open because neither of those tools knows what you actually invoiced last month. Somewhere there's a chat thread with three decisions buried in it, and an inbox quietly filling up. You've typed the same customer's name into four different boxes this week, spelled it slightly differently in two of them, and you'll probably do it again before Friday. None of this is the business. It's the scaffolding around the business — and most mornings the scaffolding wins.
Tool sprawl on the left, the same customer fragmented across each. On the right, one place where the work, the people and the numbers share what they know.
The cost nobody puts on the invoice
You can add up the subscriptions easily enough — a line item here, a per-seat charge there, the annual plan you forgot you were still paying for. That number stings, but it isn't the real cost. The real cost is quieter and harder to name. It's the customer who asked a question last week and got a slightly different answer from you than from your helper, because the two of you were looking at two different tools. It's the invoice that went out late because the deal closed in one app and the billing lived in another and nobody carried the news across. It's the half-hour every evening you spend being a human bridge between systems that were each, individually, supposed to save you time.
And then there's the part that doesn't show up anywhere: the mental tax of switching. Every time you jump from the CRM to the tasks app to the spreadsheet, you pay a small toll in attention. Do that forty times a day and you end most days tired without being able to point to a single hard thing you did. The tools didn't lie to you. Each one is genuinely good at its one job. The problem was never any single tool. The problem is that you're the integration layer, and you're doing it by hand, all day, in your head.
So what would it feel like to stop bridging?
Here's a question worth sitting with for a second. What if the customer's name only existed in one place? Not synced, not copied overnight by some fragile connector you're afraid to touch — just one record that the sales view, the task list, the invoice and the support note all looked at directly. You change the email once. It's changed everywhere, because there was never a second copy to fall out of step.
That's the shift SMAIVIZ is built around. It isn't another app to add to the pile. It's one platform, one login, and underneath it a set of modules that share the same data instead of guarding their own. Your CRM lives there. So does the work and the tasks. So do your books, your customer data, and the automation that ties it together. They aren't five products wearing a matching coat of paint — they genuinely sit on the same ground, which is why nothing has to be re-typed and everything already connects.
The pieces, and why they're better together
You can pick up just the part you need first and grow into the rest. Most people start where the pain is loudest and notice, a week or two in, that the other corners have quietly stopped being a chore.
Orbit holds your customers and deals — the relationships, the pipeline, the next thing you promised someone.
Kriya is where the work and the tasks live, so "follow up with that account" is a thing that actually has an owner and a date, not a sticky note.
Hive treats your people and your AI agents as one team, so help — human or machine — shows up in the same place.
Artha keeps your books, sitting right next to the deals that created the invoices instead of in a tool that's never heard of them.
Prism keeps your customer data clean and whole, so one person isn't three half-records.
Flux handles the automation — drawn on a visual canvas or just described in plain English — so the handoffs you do by hand can quietly do themselves.
The point isn't the list. The point is what happens between the items. A deal closes in Orbit, and because the books in Artha look at the same record, the invoice has everything it needs. A new customer arrives, and the task to welcome them lands in Kriya already attached to the right person. You stop carrying news from one room to another, because it turns out they were the same room all along. And because you bring your own model, the AI woven through it works the way you choose, on your terms.
One place doesn't mean less control
There's a fair worry when everything connects: does it mean things start happening on their own, in ways you can't see? It shouldn't, and here it doesn't. The sensitive moves — sending money out, emailing a whole list, changing something that's hard to take back — wait for a human to say yes. Automation can tee up the work and present it neatly, but the decision stays with a person.
And because it all lives in one place, there's a clear trail of what happened, who approved it, and when. Not as a compliance chore, but as plain peace of mind. When you wonder "did that invoice actually go out, and who sent it?" the answer is right there, in the same system as everything else. You're not piecing the story together from five tools' worth of half-memories anymore.
Questions you're probably asking
Do I have to move everything over at once?
No. Most people start with the one module that hurts most — usually the CRM or the tasks — and let the others come online when they're ready. Because everything in SMAIVIZ shares the same data from the start, adding the next piece doesn't mean another migration or another connector; the new module simply already knows about the customers and work you've got.
Won't one big platform be worse at each individual job than a specialist tool?
Each SMAIVIZ module is built to stand on its own for its job. What you give up isn't quality — it's the gaps between tools. The CRM, the books, the tasks and the automation are all serious about their own work, and they also happen to be looking at the same data, which a specialist tool on its own can never offer.
Is the AI going to do things without me knowing?
No. Anything sensitive pauses for a human to approve, and every action leaves a clear, time-stamped trail you can read back. The AI helps draft, sort and tee things up inside SMAIVIZ — but you stay the one who says yes, and you bring your own model so it works on your terms.
What does "shared data" actually save me?
The re-typing and the drift. A customer exists once across SMAIVIZ, so a change to their details shows up everywhere at once and there's no second copy to fall out of sync. You stop being the human bridge that carries information between apps by hand.
Imagine the version of your morning where you open one thing. The deal that moved, the task it created, the invoice it's owed, the person behind all three — same place, same truth, no bridging. The scaffolding fades into the background and the actual business steps forward. That's the whole idea: less time tending your tools, more time running the thing they were supposed to serve. Start free at smaiviz.com.
It's a little after eleven at night, and you're not asleep. You're scrolling back through the day in your head, landing on the same thought you keep circling: somewhere out there, a tool you set up this morning is still running. It's sending things. Touching records. Maybe drafting a reply to a customer who matters. And you genuinely don't know — not for certain — exactly what it did or what it's about to do next. You wanted the help. You asked for the speed. So why does it feel like you handed your name to a stranger and walked away?
The routine work flows on its own. The moment something sensitive comes up, it waits for you — and every step leaves a record.
Here is the trap a lot of careful leaders fall into. You hear what AI can do, and part of you leans forward — the part that's tired of the same repetitive work eating your team's afternoons. Then another part pulls hard the other way. Because the stories that stick aren't the ones about saved hours. They're the ones about the automated message that went out with the wrong tone to the exact wrong person. The change nobody noticed until a customer asked about it. The payment that moved a beat before anyone meant it to.
So you do the thing that feels responsible: you keep AI on a short leash, or off entirely. You let it draft, but you retype everything yourself. You let it suggest, but you do the actual work by hand. And it helps a little — but you've quietly given back most of the speed you wanted, because the only way you trust the machine is to do its job over again behind it.
The quiet cost of doing it all yourself
That caution isn't wrong. It's just expensive in a way that doesn't show up on any invoice. It shows up in the evenings you spend reviewing things that were already fine. It shows up as the bottleneck that is you — the one person every small decision has to pass through before anything ships. Your team slows to your pace. Good ideas wait in a queue behind your attention. And the irony is brutal: you adopted AI to stop being the bottleneck, and now you're more of one than ever, because you're personally guarding the gate against a tool you don't fully trust.
Underneath all of it is one honest fear, and it deserves to be said plainly. It isn't that AI will be useless. It's that it will be confident and wrong, and act on that, with your name attached. You can live with a tool that's occasionally unhelpful. What keeps you up is a tool that's occasionally unstoppable.
So here's the question worth sitting with: what if the choice was never really between speed and control? What if the problem was simply that, until now, the tools made you pick one place to stand — fully hands-on, or fully hands-off — when what you actually needed was to be hands-on about the few things that matter and hands-off about the rest?
A different shape for the deal
This is the idea SMAIVIZ is built around, and it's a quieter idea than most of what you hear about AI. The routine work — the sorting, the drafting, the follow-ups, the tidying that fills a day — the AI agents just handle. That part should be fast, and it is. But the moment something sensitive comes up, the work stops and waits for you. It doesn't guess. It doesn't proceed and hope. It pauses, shows you exactly what it wants to do, and asks.
You're the one who decides what counts as sensitive, because only you know where your real risk lives. Sending an outbound message to a customer. Moving money. Changing something that's hard to undo. Each AI agent has a clear, defined boundary — a list of what it's actually allowed to do on its own, set by you — and it can't quietly wander past it. An agent that helps with customer records doesn't get to touch your finances just because it had a thought. The leash isn't a single short tether anymore. It's a fence you drew yourself, around exactly the right yard.
And because the same idea runs through everything — your automations in Flux, your tasks in Kriya, your customers in Orbit, your people and AI agents working as one team in Hive, your books in Artha, your customer data in Prism — you only learn the rule once. One login, shared information, and the same calm promise across all of it: the routine moves, the sensitive waits.
The two things that let you actually sleep
If you strip the idea down to what makes it feel safe rather than just sound safe, it comes to two things. The first is the pause. The second is the record. Neither is glamorous. Together they're the whole reason you can hand off real work without handing off your peace of mind.
Start with the pause, because it's the part you feel first. When an AI agent reaches a sensitive step, it doesn't act — it puts the action in front of you and holds. You see what it intends to do before it does it, in plain language: this message, to this person; this change, to this record; this amount, moving here. You approve it and it goes. You decline it and it doesn't. Nothing sensitive ever happens on your behalf without a person — you, or someone you trust — saying yes first. The speed lives in everything leading up to that moment; the control lives in the moment itself.
Then there's the record, which is the part you feel later, usually on the day you need it most. Everything that happens leaves a clear trail. What the agent did, what it proposed, who approved it, and when — written down in order, so there's never a mystery to reconstruct from memory. When someone asks "wait, why did this go out?" you don't squint and guess. You look. The honest relief of a good audit trail isn't really about catching mistakes, though it does that. It's that you stop having to hold the whole day in your head. You can let go of a task and still answer for it, because the system remembers so you don't have to.
And if you'd rather the thinking itself run on an AI model you already trust, you can bring your own. The judgment stays yours all the way down — which work to hand off, where to draw each line, whose yes is required, and even whose intelligence does the reasoning.
Picture the same eleven-o'clock moment with that in place. The tools are still running. But the thought that finishes the day isn't "what did it do?" It's "it did the small stuff, and the one thing that mattered is sitting in my queue, waiting for me, with the whole story attached." That's a very different feeling to fall asleep on. Isn't that the version of help you actually wanted — the kind that makes you bigger, not the kind that makes you nervous?
Questions you're probably asking
Won't pausing for approval just slow everything back down?
Only the sensitive steps pause — and you decide which ones those are. The routine work keeps moving on its own, so you're not reviewing everything, just the handful of moments that genuinely deserve a second look. With SMAIVIZ, the goal is to remove you as the bottleneck on the small stuff while keeping you in the loop on the things that matter.
How do I stop an AI agent from doing something it shouldn't?
You set what each agent is allowed to do, and it can't go past that boundary. An agent scoped to one kind of work can't quietly reach into another. In SMAIVIZ those limits are yours to draw and change, so the fence always matches your comfort, not someone else's default.
If something does go wrong, can I see exactly what happened?
Yes. Every action and approval is logged in order — what was done, what was proposed, and who said yes. When a question comes up later, you look at the record instead of reconstructing it from memory. That clear audit trail is a built-in part of how SMAIVIZ works, not an add-on.
Can I use the AI model I already trust?
You can bring your own model. The reasoning can run on the AI you prefer, while the approvals, boundaries, and audit trail around it stay the same. SMAIVIZ is designed so the control layer is yours regardless of which model is doing the thinking.
You don't have to choose between moving faster and staying in control. You only had to until the tools caught up to how careful people actually want to work. Let the AI carry the routine, keep your hand on the few things that matter, and let the record hold the rest. Start free at smaiviz.com.
It is a Tuesday afternoon when the email arrives. The subject line is polite, almost friendly: "Quick question about my data." A customer you barely remember wants to know what you hold on them, where it is kept, and whether you can delete it. You read it twice. Then you sit very still, because the honest answer is that you are not entirely sure. Their details are in a spreadsheet, yes — but also in your inbox, in an old invoicing tool, in two chat threads, and probably in an export someone made last spring that nobody ever cleaned up. You could probably find most of it. Probably.
When a customer asks "what do you have on me?", the answer should take minutes, not a frantic afternoon.
For a while, that uncertainty is survivable. You answer the email as best you can, you make a note to "sort the data situation out one day," and you move on. But the feeling lingers — that low hum of anxiety that sits underneath a growing business. You are not careless. You care a great deal. It is precisely because you care that the not-knowing is so uncomfortable.
The quiet cost of not knowing
The trouble with scattered data is not usually a dramatic breach. It is the slow, daily erosion of confidence. Every time someone on your team copies a list into a new tool, every time a contractor gets handed a login "just for now," every time an export lands in a downloads folder, your picture of where customer information actually lives gets a little blurrier. Nobody decided this. It accumulated, one reasonable shortcut at a time.
And so a few questions start to feel genuinely hard to answer. If a customer asked you to delete everything, could you honestly say you got it all? If a careful client wanted to know who on your team had viewed their account, would you have any record? If a regulator — or just a thoughtful prospect doing due diligence — asked where your data is physically kept, would you know the answer or have to go and find out?
None of these are exotic. They are the ordinary questions of doing business with people who, quite reasonably, want to trust you with their information. The cost of not being able to answer them is rarely a fine. More often it is something quieter: a deal that cools, a customer who never quite relaxes, and your own steady, draining awareness that you are operating on hope rather than certainty.
The turning point is a single question
Here is the question worth sitting with: not "are we compliant?" — which is abstract and easy to defer — but something far more concrete. If the most careful customer I have emailed me tomorrow and asked for everything, where it lives, and who has touched it, how long would it take me to answer with confidence?
If the honest answer is "days, and I'd still be nervous," that is not a failing of character. It is a failing of structure. Data that is scattered cannot be accounted for, no matter how conscientious you are. The fix is not to try harder. It is to give your customer information one home — a place where being able to answer those questions is simply how the system works, rather than a heroic effort you mount each time.
One home, with the answers built in
This is where SMAIVIZ comes in, and it is worth being plain about what that actually means for you. SMAIVIZ brings the parts of your business that touch customer data — your customer records in Orbit, the way you tidy and de-duplicate them in Prism, the work and conversations around them — under one login with shared data. Instead of the same person existing as five half-versions across five tools, they exist once, in one place you can actually see.
That single home is what makes the hard questions easy. Because the information lives in one accountable place, the awkward requests stop being emergencies:
"What do you have on me?" You can find a customer and see what is held, instead of reconstructing it from memory and old folders.
"Please delete me." You can honour an access or deletion request properly, knowing you are working from the real record and not a stray copy you forgot existed.
"Where is my data kept?" You choose the region where your data is stored, so you can answer plainly when someone asks — no guessing, no shrug.
None of this requires you to become a privacy lawyer. It requires the underlying setup to be honest by default, so that doing the right thing is the path of least resistance rather than a project you keep postponing.
Bringing AI in without giving up the keys
More of us now want AI helping with this work — drafting the reply to that data request, summarising an account, spotting duplicates a human would miss. That is genuinely useful. It also raises a fair worry: if an assistant is reading your customer data, whose hands is it passing through, and what is being done with your keys?
The SMAIVIZ answer is that you bring your own model. Your AI keys stay yours. The assistant helps inside your workspace, on your terms, without quietly becoming one more place your customer data has wandered off to. The help is real, and the boundary stays where you put it.
Staying in control of the sensitive moments
Convenience and control usually feel like opposites. The interesting part is when they stop being a trade-off. In SMAIVIZ, the routine things move quickly, but the sensitive ones — exporting a batch of records, deleting a customer's data, anything with real consequences — wait for a human to approve them. A person looks, understands, and says yes before it happens. Automation does the tedious work; a human keeps the judgement.
And underneath all of it runs a clear record of the important actions: what was changed, who did it, when. Not so you can police your team, but so that when someone asks "who touched this account?" you have an answer that is a fact rather than a hope. That record is the difference between believing you are accountable and being able to show it.
Imagine the next version of that Tuesday email. The same polite subject line, the same careful customer. Only this time you open the record, see exactly what you hold, confirm where it lives, and handle their request the same afternoon — with a calm you had almost forgotten was possible. What would it be worth to never dread that email again?
Questions you're probably asking
If a customer asks me to delete their data, can I actually do that properly?
Yes. Because SMAIVIZ keeps customer information in one accountable place rather than scattered across tools, you can find what you hold and honour an access or deletion request from the real record — not a half-remembered set of copies. Sensitive deletions wait for a human to approve, so nothing happens by accident.
Where is my data actually kept?
In the region you choose. SMAIVIZ lets you decide where your data is stored, so when a customer or a careful prospect asks the question, you can answer plainly instead of going to find out.
If I use AI on my data, who ends up with my keys?
You do. SMAIVIZ is built around bringing your own model, so your AI keys stay yours. The assistant helps inside your workspace without your customer data quietly drifting somewhere new.
How would I prove who touched a particular account?
SMAIVIZ keeps a clear record of important actions — what changed, who did it, and when. When someone asks, you have a factual answer for an audit or a careful customer, rather than relying on memory.
Will all this control slow my team down?
The everyday work stays fast. Only the genuinely sensitive actions pause for a human approval. You get speed where it helps and a checkpoint exactly where it matters.
You do not have to carry that low hum of not-knowing forever. Give your customer data one honest home, decide where it lives, keep a record you can stand behind, and let a human stay in charge of the moments that matter. The next careful question becomes one you can answer in minutes. Start free at smaiviz.com.
You signed up with the best intentions. You poured a cup of something warm, opened the workspace, and then it just sat there — clean, white, waiting. A blank canvas is supposed to feel like possibility, but more often it feels like a question you don't quite know how to answer. Where do you even begin? You toggle between tabs, read a little, close the laptop, and tell yourself you'll figure it out properly later. If "later" has a way of never arriving, you are not lazy and you are not behind. You are simply standing at the start of something, which is the hardest place to stand.
You don't conquer the whole platform on the first morning. You take four small steps, and each one stands on its own.
Here is the reassurance worth tattooing somewhere visible: you do not have to set up everything to get something out of SMAIVIZ. The whole platform lives behind one login with shared data, which means every step you take quietly makes the next one easier — but no single step depends on you having finished the others first. Add one customer and the platform is already useful. Invite one teammate and it is already shared. You are not building a cathedral this week. You are laying four small stones, and any one of them is enough to stand on.
So put the pressure down. Let's walk through the first few days the way a patient friend would walk you through it — one beat at a time, with permission to stop after any of them.
Step one: add your customers
Start with the thing you already know best — the people you serve. Open Orbit, the part of SMAIVIZ where your customers live, and add just one. A real one. A name, an email, maybe the last conversation you had with them. That's it. Resist the urge to import a giant spreadsheet and reconcile every field; you can do that later, and honestly you'll do it better once you've felt how the place works.
What you'll notice is that the empty workspace stops feeling empty the moment a familiar name appears in it. Suddenly this isn't an abstract tool — it's a place where your actual business has a face. Add two or three more if it feels good. Each one is a small win, and small wins are the entire trick to surviving a blank canvas. When you're ready, Orbit will happily take the whole list. But today, one is plenty.
Step two: invite your team
A tool you use alone is a notebook. A tool your team shares is a workplace. So the next small step is to invite the people you work with — even one of them. When you do, you'll meet Hive, the part of SMAIVIZ where your team lives. And here's the gentle surprise: in Hive, people and AI agents sit on the same team roster. The humans you trust and the AI helpers you'll build later are managed in one place, side by side, like colleagues with different strengths.
You don't have to think hard about that yet. For now, just bring one real person in so the customer you added a moment ago isn't yours alone — it's the team's. Shared data behind one login means they see what you see, without a single export or forwarded attachment. That quiet click is the moment your workspace becomes a shared room instead of a private corner.
Step three: build your first AI assistant
This is the step most people expect to be intimidating, and it's the one that tends to surprise them most. An AI assistant in SMAIVIZ — an agent — is just a helper you describe in plain language and point at a small, real job. You don't write code. You tell it what you want it to be good at.
Pick something small and a little tedious. Maybe an assistant that drafts a friendly reply to a new enquiry. Maybe one that reads a long thread and hands you a three-line summary. Maybe one that does a quick bit of research before a call so you walk in prepared. Give it a name, describe its job, and try it once on something real. You bring your own model, so the assistant runs on the AI you already trust rather than something chosen for you. The first time it hands back something genuinely useful, the blank canvas finally tips over into a workshop. Doesn't it feel different when the tool starts doing a little of the work with you?
Step four: automate one piece of busywork
By now you have customers, a teammate, and a helper. The last small step is to take one repetitive chore off your own plate. That's Flux, the part of SMAIVIZ that handles automation. You can describe what you want in plain English — "when a new enquiry comes in, draft a reply and let me approve it" — or lay it out on a visual canvas if you'd rather see the shape of it.
Choose the dullest recurring thing you do. The follow-up you always forget. The little note you copy into the same place every time. Automate that one, and only that one. The point isn't to wire up your whole business in an afternoon; it's to feel the relief of a chore that now happens without you. One automation that actually fires is worth more than a grand plan you never finish. And once you've felt it, you'll start spotting busywork everywhere — but again, one at a time.
And when you're ready for more, the rest of the platform is right there under the same login: Kriya for work and tasks, Artha for your finances, Prism for keeping your customer data clean and clear. There's no new account, no new password, no fresh blank canvas to dread. Just more rooms in a house you already live in.
Stay in control the whole way
A fair worry, especially in week one: if assistants are drafting and automations are firing, who's actually steering? You are. Anything sensitive waits for a human approval before it happens — the assistant proposes, you decide. And every meaningful action leaves a clear audit trail, so you can always look back and see what happened, who set it in motion, and why. You're not handing over the wheel. You're getting a very capable co-driver who always asks before the turns that matter.
Questions you're probably asking
Do I have to do all four steps in the first week?
Not at all. Each step in SMAIVIZ stands on its own — add one customer and you've already got something useful. Because everything shares one login and the same data, finishing one step quietly makes the next easier, but none of them depend on the others. Do one, breathe, come back for the next.
What if I get stuck and don't know where to start?
Start with what you know best: add a single real customer in Orbit. That one small win turns the blank workspace into a place where your actual business lives, and the next step always feels lighter once the first is done.
Do I need to be technical to build an AI assistant?
No. In SMAIVIZ you describe an assistant in plain language and point it at a small job — drafting a reply, summarising a thread, doing quick research. No code. And it runs on the AI model you already trust, because you bring your own.
Will automations do things without my say-so?
Not on anything that matters. SMAIVIZ keeps a human approval on sensitive actions, so an automation or assistant proposes and you confirm. Every meaningful action is recorded in a clear audit trail, so you always know what happened and why.
Is this really one tool or several I have to learn?
One. Orbit, Hive, Flux, Kriya, Artha and Prism all live behind a single login with shared data. You learn one place, not five — and you can grow into the other parts whenever you're ready, with no new account to set up.
That's the whole secret to a first week that actually sticks: not a heroic setup marathon, but four small steps you can take on four unremarkable days, each one leaving you a little less alone with the work. The blank canvas was never asking you to fill it all at once. It was only ever asking for the first stroke. Make it small, make it real, and let the rest follow. Start free at smaiviz.com.
