Trying to build some authentication protocols into it in collaboration with a bunch of Stanford folks. Very fun space!

📄 Read the full paper (examples, motivations, legal insights, technical designs, and more): arxiv.org/pdf/2501.09674

Amazing to collaborate with Samuele Marro, Thomas Hardjono, Robert Mahari, @cedricwhitney.bsky.social, Dazza Greenwood, Alan Chan, Alex Pentland, and many others providing feedback.

We’ve even tackled natural language permissions!

Our framework translates human-readable instructions into auditable, structured access controls.

Think: "Only read project files, avoid financial folders" → enforceable, machine-readable rules. 🚀

Delegation isn't just technical. It's crucial for:

⋅ Protecting human-only spaces online 🌐
⋅ Preventing misuse in high-stakes domains (finance, healthcare) 🏥💳
⋅ Enabling safe, auditable agent collaboration 🛠️

Our framework builds on existing web standards (OAuth 2.0, OpenID Connect) to:

✅ Identify agents uniquely
✅ Limit permissions & scope
✅ Maintain audit trails
✅ Ensure human oversight

All while maintaining compatibility with web infrastructure. 🌐

Autonomous AI agents are growing fast—how do we ensure they act responsibly, securely, and transparently?

OpenAI Operator isn't good enough...

🔑 Authenticated Delegation is our answer:
A secure way to delegate authority while maintaining accountability and control.

The cursor team really can't catch a break

This is the best abstract I've perhaps ever seen...

I always think not eating until the feast is a smart move until I arrive and someone starts handing me French 75s