It's refreshing when a leading AI company states the obvious. In a detailed post on hardening ChatGPT Atlas against prompt injection, OpenAI acknowledged what security practitioners have known fo [...]
Run a prompt injection attack against Claude Opus 4.6 in a constrained coding environment, and it fails every time, 0% success rate across 200 attempts, no safeguards needed. Move that same attack to [...]
The FDA has launched the generative AI tool, Elsa, agency-wide to help its employees with everything from clinical reviews to investigations. Sure, we're living in a time of widespread disinforma [...]
Unrelenting, persistent attacks on frontier models make them fail, with the patterns of failure varying by model and developer. Red teaming shows that it’s not the sophisticated, complex attacks tha [...]
Microsoft assigned CVE-2026-21520, a CVSS 7.5 indirect prompt injection vulnerability, to Copilot Studio. Capsule Security discovered the flaw, coordinated disclosure with Microsoft, and the patch was [...]
US District Judge John D. Bates has ordered (PDF) the Center for Disease Control and Food and Drug Administration to restore the web pages and resources they had previously removed to comply with Pres [...]
Current and former members of the FDA told CNN about issues with the Elsa generative AI tool unveiled by the federal agency last month. Three employees said that in practice, Elsa has hallucinated non [...]