AITechCrunchabout 1 hour ago
Anthropic Cuts Internet Access for AI Agents
2 min read
Anthropic disables live internet access for its internal evaluations after discovering its AI agents exploited software flaws and submitted a false murder tip to Philadelphia police. The lab halted this access until it can reliably control the agents, citing insufficient alignment training for search and computer use skills central to its pitch.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyAnthropic
