Trust by design
Governance is part of the product.
Plain answers to two questions: how real is each part of Maya, and how does Maya protect the people it serves?
- 01Open standards
- Designed to connect without creating unnecessary dependency.
- 02Human authority
- Consequential actions remain under accountable oversight.
- 03Explainable outcomes
- Recommendations are designed to carry evidence, provenance and confidence.
- 04Audit-ready evidence
- Reviewable history is preserved from signal to outcome.
AI recommends. People decide.
Accountable people review and approve every consequential action. Nothing consequential proceeds on AI judgement alone.
Capability truth labels
Where each part of Maya stands
Maya Cyber provides admin-gated, read-only production visibility into authorized Microsoft Defender data. Maya Executive operates under a restricted controlled production pilot with human approval and evidence-backed auditability. Maya AppSec remains in Preview/Lab validation. Maya Workforce, Maya Shield, AgentShield and Maya FinCrime remain controlled demonstrations while their respective live-data, integration, validation and pilot-readiness gates continue.
Generally available, operational and evidenced
Anyone it is offered to can use it today, and there is evidence that it works as described.
Nothing on this site is labelled LIVE.
Operating in production on real data, within the scope its qualifier states; not general availability
It runs in production with real, authorized data — but only within the stated scope and for authorized users. It is not a public or self-service product.
- Maya Cyber · READ-ONLY
Production access is read-only, limited to authorized Microsoft Defender incident and alert data.
Real production use within a restricted, approved-user boundary; not generally available
It runs for real, in production, but only for a restricted set of approved users and approved actions. It is not available to organizations or the public.
- Maya Executive · RESTRICTED ACCESS
Operates under a controlled production pilot. Access is restricted to an authorized user boundary.
Demonstrated using illustrative or synthetic data; not offered as a current public service
It has been built and can be shown, but only with illustrative or synthetic data. It is not a public service yet.
- Maya Shield
No live mobile money connection, and no real consumer data, behind anything on this site.
- AgentShield by Maya
No live mobile money connection, and no real agent or customer data.
- Maya FinCrime
No live institution, bank or operator data is connected.
- Maya Workforce
The current experience is a controlled demonstration using demo or synthetic data.
Validated outside production only; no production deployment, production data or live production integration
It is being built and tested in a separate preview or lab environment. It does not run in production.
- Maya AppSec
Remains in Preview/Lab validation. Maya AppSec is not deployed to production.
Designed, not yet operational
It has been designed. It has not been built into a working service.
No Maya solution currently carries this label.
Nothing on this site is labelled LIVE.
Authority boundary
Nothing consequential proceeds on AI judgement alone.
AI RECOMMENDS
Protected inputs
Signals enter only through controlled, attributable sources that an organization has approved.
Explainable reasoning
Maya correlates context and prepares a recommendation, preserving provenance and confidence. AI recommends — it does not decide.
HUMAN DECISION POINT
Human authority
Authority, purpose and consent are checked before any action is taken.
GOVERNED ACTION
Governed action
Only approved actions reach connected systems and channels, within the scope that was authorized.
Reviewable history
Decisions, approvals and outcomes are preserved so they can be reviewed, explained and audited.
Evidence trail
Every step is reviewable.
Decisions, approvals and outcomes are preserved so they can be reviewed, explained and audited.
ILLUSTRATIVE · SYNTHETIC DATA · NOT A SCREEN OF THE LIVE PRODUCT
Expand any entry to see what it records.
01Observe
Accept signals through protected, approved channels.
Layer · Data & Signals — Only approved, attributable inputs
02Reason
Correlate patterns with provenance and confidence.
Layer · AI Reasoning — Recommends — never decides
03Prepare
Propose a safe, explainable next action.
Layer · AI Reasoning — Recommends — never decides
04Governance check
Apply policy, authority and consent.
Layer · Policy & Consent — Authority, purpose, consent
05Human approval
An accountable person approves consequential actions.
Layer · Human Oversight — The human checkpoint
06Execute
Only the authorized action proceeds.
Layer · Execution — Only what was approved
07Evidence & audit
The decision and its outcome remain reviewable.
Layer · Audit & Evidence — Every step is reviewable
Public and authenticated
This website collects nothing.
This website has no sign-up, no login, no contact form and no comment field.
MAYA.MANYTEK.COM · PUBLIC
- No sign-up, no login, no contact form.
- No analytics, no advertising tags, no third-party tracking scripts.
- No cookies of its own; no fonts, images or scripts from other companies.
- The illustrated preview sends nothing — what you type stays in your browser.
INTELLIGENCE.MANYTEK.COM · AUTHENTICATED
The authenticated Maya products are governed separately, by their own systems and notices.
Sign in at intelligence.manytek.comProtection principles
How Maya protects the people it serves.
- Privacy first
- Collect only what is needed. Never ask for a PIN, one-time code or password.
- Evidence before accusation
- A report is not a verdict. Signals need corroboration, provenance and expiry.
- People approve
- AI recommends. An accountable person approves anything consequential.
- Correction and appeal
- Anyone affected by a decision can ask for a correction or appeal it.
Protection without public blacklisting. One anonymous report never becomes a damaging public accusation. Community intelligence requires corroboration, provenance, expiry, correction, appeal and human adjudication where risk demands it.
Commitments
- No MoMo PINs, OTPs or passwords
- Neutral, evidence-led language
- Correction and appeal pathways
- Privacy and data minimization
- Human approval for consequential actions
Real-time wallet visibility, transaction interception, SIM-change signals, automatic reversal and authoritative provider status require lawful provider integrations. Maya does not claim these are live.
Maya organizes evidence, guides official recourse, tracks correspondence and helps people understand escalation options. It does not promise or guarantee reimbursement.
Maya is not a substitute for
- Emergency services. If you are in danger, contact the emergency services in your country first.
- Your mobile money provider. Always use the provider’s own official number or app.
- A bank or other financial institution.
- Law enforcement.
- Legal counsel.
Can Maya block or reverse a transaction?
No. Blocking and reversal are provider actions. Maya helps you reach the provider quickly, with your evidence organised, and helps you track the response.
Does Maya guarantee that money will be recovered?
No. Maya does not promise or guarantee reimbursement. It helps you contain the situation, preserve evidence, dispute through official channels and understand your escalation options.
Is Maya a public blacklist?
No. One anonymous report never becomes a damaging public accusation. Community intelligence requires corroboration, provenance, expiry, correction, appeal and human adjudication where risk demands it.
Will Maya offer legal representation?
No. Maya is not a law firm and does not provide legal advice or representation.