I’ve joined the U.S. AI Safety Institute. Building government capacity for safe AI innovation is a dream job. The team has achieved so much, and I look forward to accelerating the work as a Senior Advisor. If you may be able to help, please don’t hesitate to get in touch.

RT @emollick: It seems like there is not enough of a policy response to the fact that, with 57M miles of data, Waymo’s autonomous vehicles….

You can find the full workshop read-out and an email address for feedback on the CAISI website:

A shared vocabulary for AI agent tools and tool use supports information sharing and risk management in AI development and use. The ecosystem can build on these initial taxonomies to meet specific needs.

The second AI agent tool taxonomy considers constraints: how can use of the tool be controlled?

The first AI agent tool taxonomy considers functionality or affordances: what does the tool enable the agent to do?

CAISI and @NIST hosted a workshop on AI agent tool use with 140 experts. The goal: develop a shared vocabulary for agent tools. Today we published the result, with two example taxonomies:.

RT @mkratsios47: A year ago, policymakers in DC were debating a ban on open weight models. From day one, the Trump administration changed….

RT @DavidSacks: 7/ Acknowledging sensible risk mitigation. Although we want a light touch on AI regulations, President Trump made clear t….

I’m at #icml25 this week, presenting a @taig_icml poster on societal resilience indicators. The paper with Milan Gandhi, @OCLarter, and @rebershak was inspired by our work on Microsoft’s frontier governance framework. Reach out if you’d like to chat in Vancouver.

RT @simonw: Quitting programming as a career right now because of LLMs would be like quitting carpentry as a career thanks to the invention….

RT @CommerceGov: RELEASE: Statement from U.S. Secretary of Commerce Howard Lutnick on Transforming the U.S. AI Safety Institute into the Pr….

RT @VaticanNews: Pope Leo XIV explains his choice of name:. ". I chose to take the name Leo XIV. There are different reasons for this, bu….

RT @deanwball: How should the federal government prioritize its AI R&D investments over the next 3-5 years?. If you have thoughts, we want….

RT @EdelmanBen: 1/ Excited to share a new blog post from the U.S. AI Safety Institute!. AI agents are becoming increasingly capable. But th….

I’ll be at NeurIPS Thursday through Sunday. I’m presenting a poster on agent measurement at the SoLaR workshop on Saturday. Please reach out if you want to chat!.

RT @avicgoldfarb: NBER Digital Economics and AI PhD student workshop. Deadline in 10 days.

Grateful today to be cancer free. It was surprisingly reassuring to have systems like Claude to help me interpret medical reports along the way. Happy Thanksgiving!.

ICYMI, much progress coming from the international convening last week:

RT @kevin_klyman: The US AI Safety Institute is hiring! Looking for experts in designing/implementing evaluations for the capabilities/safe….