Kevin Wei

After 11 months of work, we proudly announce Third Opinion: A free of charge expert consultation service for frontier AI professionals. To help you clarify if what you're seeing is cause for concern. Anonymous, without sharing confidential information 🧵 tinyurl.com/2rbk2w59

🚀Come join my team at RAND! We’re looking for research leads, researchers, & project managers for our compute, US AI policy, Europe, & talent management teams. All teams have urgent, important work to do & broad options for the future. Some roles close July 27⏰

🚨 AI Evals Crisis: Officially kicking off the Eval Science Workstream 🚨 We’re building a shared scientific foundation for evaluating AI systems, one that’s rigorous, open, and grounded in real-world & cross-disciplinary best practices👇 (1/2) evalevalai.com/research/2025/…

As AI agents near real-world use, how do we know what they can actually do? Reliable benchmarks are critical but agentic benchmarks are broken! Example: WebArena marks "45+8 minutes" on a duration calculation task as correct (real answer: "63 minutes"). Other benchmarks…

📢 Last Call for Applications! Apply by May 31 to join one of our three in-person events this summer: 📆 Summer Institute on Law and AI: July 11-15, Washington, DC-Area 📆 Workshop on Law-Following AI: August 6-8, Cambridge University, UK 📆 Cambridge Forum on Law and AI: August…

Really excited to share my first ever paper! “Third-party compliance reviews for AI safety frameworks” 🚀 See below for more ⬇️

"By establishing state data commons, policymakers can help ensure that AI’s benefits extend to all communities, advancing the promise of technology while preserving democratic values and local autonomy," write @kevintfrazier and @kevinlwei.

Indeed. If / when they do this at Commerce they'll wipe out the entire US AI Safety Institute. Not smart @elonmusk

📣 Announcing the AI Agent Index AI agents are growing in number, capabilities, and impact. In response, we introduce the first public resource documenting the technical and safety features of deployed agentic AI systems. aiagentindex.mit.edu

Whether 2025/6/7/8/etc is the year of AI agents, we will need tools to help us unlock their benefits and manage their risks. Our new research agenda proposes agent infrastructure as an answer to this challenge 🧵

1/ New preprint! Findings from our new survey capture the views of local US officials on the impacts and governance of AI before and after the release of ChatGPT. Authors: @SophiaHatz4 @kevinlwei @baobaofzhang and myself, fielded by @CivicPulse.