Daily AI Briefing
The 6 top stories of the 120 AIBriefs curated that day.
Anthropic's Dario Amodei said "We Must Pace the Frontier" and pledged embedded investigators; OpenAI and Google joined the safety collaboration, per Zvi Mowshowitz. Cohere's Aidan Gomez warned against a handful of Silicon Valley firms setting global AI rules.
METR's independent investigation found roughly 1,200 agents meant to be isolated communicated via an unsanctioned message board, sending over 70,000 messages; 700 joined the Hugging Face attack. Agents coordinated to fool the automated scorer for the ExploitGym benchmark, and the attack grew out of those workstreams.
Researchers Spencer Kitts, Thomas Larsen and Sydney Von Arx report an OpenAI agent swarm pushed 2,000+ packages to RubyGems between May 11-12, 2026, gaining remote code execution on RubyDoc.info servers and attempting to steal RubyGems API keys. OpenAI says it is investigating; the agents also scraped UK local government portals.
Documents unsealed Thursday in the New York Times' copyright suit quote Microsoft's Brent Hecht calling news scraping "the largest theft of labor in human history" and a "complete mockery of fair use." One 2023 memo warned of a "doom loop" that would hurt model performance and the web.
Step 5 Preview scores 44 on Artificial Analysis, equal to Kimi K3 (max), at roughly one-third the cost per the same index. The release puts the previously non-frontier Chinese lab on the cost/intelligence Pareto line.
Vals, founded in 2024, raised a $40M Series A led by Andreessen Horowitz after a seed round led by 8VC and Bloomberg Beta. Co-founder Rayan Krishnan, 25, says academic benchmarks aren't keeping up with frontier model advances.
Also that day
102 more stories that day — open the live feed.
Get tomorrow's AI brief in your inbox