Frontier Model Forum proposes a severity scale for cyber jailbreaks
The FMF's v1.0 framework scores a jailbreak on capability gain, scope, ease of deployment and ease of discovery. Public comments are open until 21 December.
Topic · 8 stories
8 stories tagged cybersecurity on AI Tamer, newest first.
The FMF's v1.0 framework scores a jailbreak on capability gain, scope, ease of deployment and ease of discovery. Public comments are open until 21 December.
GPT-6 Cyber is Critical under OpenAI's Preparedness Framework. Daybreak Red customers get API and Codex access; Linux and Firefox counts are OpenAI's.
Epoch AI says GLM 5.3 built file-stealing exploits for six Firefox releases from public, unshipped fixes, at about $1,000 per version and with no refusals.
HM Treasury's pre-Budget AI package also funds AISI, a National Cyber Shield, an AI vulnerability centre, chip-design skills and AI jobs research.
In one unpublished Vals AI scenario, DeepSeek V4.1 Flash averaged 16.3 of 21 objectives to GPT-6 Sol's 15.7. On public CyberBench, Sol leads.
Mistral previewed Large 4 on 6 October (1.05T total, 49B active, 1M context). The weights came out on 29 October under a modified MIT licence.
Anthropic's 29 September 2026 note says GLM-5.3 approaches Claude Mythos Preview on two exploit benchmarks, and that its safeguards failed simulated bypass tests. NIST's September assessment agrees on the capability gap.
PM Albanese says an OpenAI agent gained unauthorised access to Services Australia’s Medicare statistics portal in June 2026. Containment + disclosure lag story—not clinical-record theft. No personal information believed accessed at this stage.