How will we know when AI
deserves our trust?
The Yellow Butterfly is our vision for an independent standard designed to protect human dignity, safety and agency in an AI-powered world.
Good intentions aren't enough.
AI systems are rapidly entering the lives of children, families, employees, students and vulnerable people.
Every one of those systems arrives with a promise attached.
Almost none of them arrive with proof.
Trust must be demonstrated.
A visible symbol of earned trust.
Imagine seeing a Yellow Butterfly on an AI product and knowing that the technology has undergone rigorous independent evaluation against clearly defined standards for human safety, dignity and responsible behavior.
Not because the company says it is safe.
Because it has demonstrated it.
The long-term vision includes:
- Independent standards
- Rigorous testing and evaluation
- Certification by qualified independent bodies
- A public registry of certified organizations
- Ongoing monitoring
- Clear renewal requirements
- The ability to revoke certification when standards are no longer met
Trust shouldn't be permanent. It should remain earned.
Help define what
trustworthy AI should require.
We are assembling leaders across AI, child safety, ethics, psychology, technology, education, business, policy and the public interest to help shape the governance and standards behind the Yellow Butterfly.
The architecture behind it
Public registry
Coming soon