REAL ANSWERS FOR A BRIGHTER TOMORROW.
YOUR KNOWLEDGE. YOUR AI. YOUR FUTURE.
Ask anything. Find real answers.
Search the world with trusted evidence, then continue with your Personal AI.
Q Search · Search results & advertising →
PEOPLE + PURPOSE + PROGRESS + A BRIGHTER TOMORROW
MyAI
Your avatar is a manifestation of your PAI. Appearance does not grant authority.
Avatar settings
Memory vault
Memories are separate objects with their own permission and lifecycle.
Permissions
The AI may propose actions, but the control plane decides what is allowed.
Your AI credits
Checking protected wallet…
1 MyAI Credit = £0.01 of customer AI credit. The wallet is separate from the monthly MyAI subscription.
Add MyAI Credits
Choose a top-up. Credits are added securely after your payment is verified.
Wallet activity
Loading wallet ledger…
Personal AI Pro
Your recurring subscription funds the MyAI platform. AI model consumption is funded separately from your MyAI Wallet.
✓ Persistent PAI identity
✓ Memory & permissions
✓ AI avatar & voice layer
✓ Protected audit
✓ Connected actions as released
Audit history
A human-readable view of consequential PAI events.
Our commitments to you
Working charter · 9 September 2026. These principles are built into MYAI; they are not editable memories.
- The Governing Principle
If the architecture cannot enforce a principle, we shouldn't claim it yet.
- Identity Persists
The Personal AI belongs to the human, not to whichever underlying model happens to be running.
- Memory Has Meaning and Lifecycle
Remembering something, knowing what kind of thing it is, and knowing when it should no longer be remembered. That's what personal AI memory should mean.
- Autonomy Has Boundaries
Autonomy is not the absence of human control. It is human control expressed through trusted boundaries.
- Authority Is Explicit
Authority should be grantable, scoped, inspectable and revocable.
- Actions Are Accountable
Important actions should leave understandable evidence of what happened.
- Human Approval Is a Real Boundary
Approval should be part of the architecture, not a reassuring sentence in the interface.
Principle → Guarantee → Evidence
Each capability must earn its claim through evidence. Aspirational capabilities must be described as such.
MYAI should never need to ask for trust that its architecture has not earned.
Trust should be inspectable.
MyAI does not ask a human to trust a promise that its architecture cannot support.
Uncertainty is exposed.
Persistence remains human-authorised.
Principle → Guarantee → Evidence
Proposed before persistence. Approved memories are kept in the human's Memory vault.
Memory meaning may be classified as fact, preference or instruction. When MyAI cannot justify a classification, it exposes that uncertainty instead.
A proposed memory does not become persistent memory until the human explicitly approves it.
Memory is not merely stored. It can be superseded or explicitly forgotten through controlled lifecycle paths.
Important actions should leave understandable evidence. Trust claims must be demonstrated by the architecture that enforces them.
If the architecture cannot enforce a principle, we shouldn't claim it yet.
MyAI should never need to ask for trust that its architecture has not earned.
App settings
Your PAI identity is derived from your authenticated account and cannot be edited in the browser.