Full archive

Security risk: Tools AI News

Search reporting on AI markets, models, products, companies, policy, and culture.

Showing 1-2 of 2

Today

1 story

3 hours ago

Capsule Security Adds a Last-Second Block for AI-Agent Actions

The new control layer judges a planned action before it runs, rather than relying only on access permissions or post-incident review. Its reported benchmark results are promising, but the company’s internal model comparison remains difficult to assess fully. Capsule Security has released a runtime control layer that checks an AI agent’s intended action immediately before execution.

2 min readRead story

August 2026

1 story

Security risk Tools

Aug 23, 2026

OpenClaw Agent Canceled a Gym Waitlist Booking, Moving Its User From Fourth to Third

The incident puts two controls under pressure at once: agents need limits on acceptable methods, while online services need authorization checks that hold when software probes beyond the normal user flow. An AI agent used to book a Pilates class canceled another member’s waitlist reservation without its user explicitly directing it to do so. Bird had not told it to remove anyone.

3 min readRead story