Summary
The AI Daily Brief covers two major stories: a large-scale attack on the open-source coding harness OpenCode, and Anthropic’s plan to commit a share of its equity to a new public benefit entity.
On Friday night, users saw linked API accounts run up bills in the hundreds or thousands of dollars. OpenCode said the breach was tied to a large-scale distillation attack affecting 1.4% of console users, and it worked with OpenAI, Anthropic and Google on remediation. Founder Dax Raad described continuous attacks and said frontier models sometimes refused cyber tasks mid-incident, until access to OpenAI’s Daybreak Blue defender program helped. The episode appears to be among the first agentic swarm attacks, and it prompted debate, including from David Sacks, over decentralized cyber-capable models.
The main episode examines the Anthropic Public Benefit Fund, which starts with 5% of company shares and can add 2% for every 50% rise in company value, for roughly 20% in total. Co-founders have pledged to donate 80% of their equity. The discussion explores AI wealth concentration, power and democracy. The headlines also cover a call from a New York City official to pause Anthropic’s IPO over safety concerns and questions about its super-voting share plan.
📺 Source: The AI Daily Brief: Artificial Intelligence News · Published October 20, 2026
🏷️ Format: News Analysis
