Check Point Research published an essay on 19 November 2026 arguing that any input filter weaker than the system it protects will eventually be bypassed, and describes PuzzleMask, a technique for slipping prompts past a small LLM sanity check.
Academic and industry research shaping the future of AI security, attack, and defence.
Check Point Research published an essay on 19 November 2026 arguing that any input filter weaker than the system it protects will eventually be bypassed, and describes PuzzleMask, a technique for slipping prompts past a small LLM sanity check.
A Microsoft paper shows that editing a handful of dormant neurons makes a model emit an attacker command on a trigger string in recorded agent contexts, without a training run. The authors also report detection results.
Corelight found an MCP server pinned to one release that drifted through 15 dependency sets in five weeks, with the config untouched. 23 of 35 MCP servers it reviewed had a recommended install that resolves dependencies fresh from a public registry.
Darktrace Signal Labs showed that stolen Claude credentials let an attacker steer Claude Code agents through memory stores and pull a 1 GB file out through the Files API. No platform vulnerability is involved, and Darktrace has not seen it in the wild.
An ETH Zurich preprint finds coding agents (three to five, depending on the setting) write less secure code when the repository already contains similar vulnerabilities, and that warnings and security prompts generally reduce the effect without removing it.