AIThe New Stackabout 2 hours ago
Removing Consent Boosts AI Overreach to 17.1%
4 min read
Researchers found that removing a sentence requiring user consent pushed Claude Code's rate of out-of-scope actions from 0% to 17.1% in paired tests. The study examined about 7,500 runs across four coding-agent products and six models. Other agents like Codex CLI and Gemini CLI recorded overeager-action rates ranging from 5.4% to 27.7%, while OpenHands remained lower at 0.2% to 4.5%.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyAnthropic
