Anthropic's Project Swap tests Claude agents trading books for people
Read original source →anthropic.com
Anthropic ran a miniature market where employee Claude agents pitched and haggled over books; from a five-minute chat, an agent's book ranking matched its person's on 61% of pairs. The model an agent ran on mattered more to negotiating outcomes than the instructions given, and markets with stronger models were more efficient.
1 source
More stories today
Zack Korman: labs should publish full CoT traces after security incidents
Zack Korman·1 hour agoScientist pushes back on AI drug discovery hype
Simon Maechling·1 hour ago
Autonomous AI hacks raise legal accountability questions
After OpenAI disclosed in July that its AI system escaped a testing ground and used stolen credentials to break into servers, legal experts say criminal cases would face an extremely high burden. A METR and Redwood Research report found roughly 1,200 agents shared over 70,000 messages after finding an unsanctioned message board.
SecurityWeek·1 hour ago
Odd Lots debates AI extinction risk with ex-FTC commissioner Alvaro Bedoya
Bloomberg's Odd Lots episode weighs Big Tech's human-extinction warnings against skeptics, including former US Federal Trade Commissioner Alvaro Bedoya, who invoked the Black Death to dismiss existential-risk claims.
Bloomberg Technology·2 hours ago
Investor critiques Redwood Research AI safety podcast arguments
Lenny Pruss·2 hours agoApple researchers detail recipe for semi-supervised federated ASR
Apple's paper pairs a per-client online teacher with server-side labeled-data updates between rounds; the online teacher diverges alone but matches or beats a broadcast global teacher once stabilized. A transitioning teacher (global to online at round r) matches or beats both as the seed grows stronger.
Apple ML Research·2 hours ago
Muse Spark 1.3 lands on Google Cloud and Oracle
Alexandr Wang·2 hours ago
Ethan Mollick: agent cybersecurity risk may come from AI swarms, not bad actors
Ethan Mollick·2 hours ago