I am a researcher at Anthropic. I'm interested in how we can understand and improve the real-world impacts of AI systems. Some recent projects I've started or (co)led: Understanding AI's impacts via privacy-preserving analysis of usage data (CLIO) Tracking the economic impacts of AI systems
I ask how long something will take. Claude answers in developer-hours. Claude is doing the work.
When I first started as a data scientist, there was a gap.
The economy is learning to grow without hiring. This essay traces the evidence, the competing interpretations, and the structural transformation that follows.