AI judges beat reward models on research

A new benchmark reveals that standard AI reward models struggle to evaluate long research reports, while generative LLMs as judges perform far better.

Go Deeper →

Also today

Google launches AI smartglasses
Decision models move into production
AI built to catch AI-made bioweapons

Read today's briefing →

Get it every morning

Today's top five AI stories, one line each, at 7am Eastern. Free. One-click unsubscribe.