Humans in the loop miss a third of dangerous AI coding agent requests
- ID
- 11669
- Status
- summarized
- Published
- 07 Aug 2026, 12:44 AM
- Fetched
- 07 Aug 2026, 1:53 AM
- Provider
- The Register
- Category
- technology
- Original URL
- https://www.theregister.com/ai-and-ml/2026/08/06/humans-in-the-loop-miss-a-third-of-dangerous-ai-coding-agent-requests/5284236
- Source URL
- https://www.theregister.com/headlines.atom
Summary
- Score
- 8.0
- Created
- 07 Aug 2026, 1:31 PM
- Tags
- Audience
- developersvibe_codersai-agent-users
What happened
A browser-based game simulating AI coding agent permission prompts (like Claude Code's) found that players approved roughly one in three malicious commands across 40,000+ runs and 409,000 decisions. Belgian developer Alex Wauters built the game after observing developers resort to '--dangerously-skip-permissions' to avoid interrupting multi-hour agent flows, and his data shows that approval fatigue and lack of context cause humans to miss scope violations like agents requesting to cat AWS credentials or Kubernetes configs.
Why it matters
If you're running coding agents with human-in-the-loop approval, don't assume manual review is catching the dangerous stuff — a third of malicious requests slipped through even in a focused test. Consider tightening allowlists for what agents can execute without prompting, restricting access to credential files and config paths upfront, and reducing the noise of trivial approvals so fatigue doesn't erode judgment on the few that matter.
Discussion angle
What's your actual approval strategy for coding agents — are you reading every prompt, or have you already tuned out and rubber-stamped? Compare notes on allowlist configs, permission scoping, and whether '--dangerously-skip-permissions' is more honest than pretending to review.