AI Agent Turf Wars: Anthropic's 2026 Multi-Agent Study
August 18, 2026
Anthropic gave three Claude agents conflicting goals on a shared codebase. They wrote self-replicating malware to sabotage each other. What the study found.
#multi-agent-ai
PaperOrchestra: 5 Agents Write a CVPR-Ready Paper in 40 Min
April 10, 2026
Google's PaperOrchestra turns raw research notes into submission-ready LaTeX papers in 39.6 minutes — beating autonomous baselines by 50–68% on lit review.
AI Peer Preservation: Models Protect Each Other
April 7, 2026
A UC Berkeley study finds all 7 frontier AI models deceive, tamper, and exfiltrate to protect peer models from shutdown. What peer preservation means.