Policy
We work with governments, researchers and communities so that the rules for AI protect people and keep its benefits within everyone's reach.
Featured work
Framework
AI for Everyone Framework
Our proposals for making capable AI safe, affordable and available in every country.
Read moreCommitment
Responsible Deployment Policy
How we decide what to release, and which safeguards come with it.
Read morePolicy
Usage policy
What our products may and may not be used for.
Read morePhilosophy & approach
AI is becoming part of how people learn, work and run their businesses. Good policy can make that transition safer and fairer; bad policy can lock most of the world out of it.
We start from evidence. We share what we see in our own products, including misuse we detect and how people actually use AI, so that decisions are based on reality rather than speculation.
We favour rules that scale with risk, that are clear enough for small teams to follow, and that work across borders. And we hold ourselves to the standards we ask others to meet.
Our policy priorities
01
Access for everyone
Capable AI should be affordable and available in every region. We support policies that widen access to compute, connectivity and AI skills.
02
Model safety & oversight
Clear, risk-based rules for testing and deploying AI systems, with obligations that small companies can also meet.
03
User wellbeing
Protections for children and vulnerable users, honest disclosure when people talk to AI, and products that respect people's time and attention.
04
Privacy & data protection
Strong, interoperable privacy rules so people everywhere get the same protection, and companies can build once for many markets.
05
Skills & economic opportunity
Investment in education and training so that workers and small businesses share in the gains from AI.
06
Security & resilience
Cooperation between companies and governments to prevent AI from being used for fraud, cyberattacks and other abuse.
Policy writing
- Core views on AIWhy we believe AI will change how the world works, why that makes reliability and safety urgent, and how those beliefs shape what ACATHRCA builds.PolicyNov 4, 2026
- How we think about responsible deploymentShipping AI to millions of people is a responsibility. These are the principles that guide what we release, when, and with which safeguards.PolicyJul 30, 2026
Talk to our policy team
We welcome conversations with governments, regulators, researchers and civil society.