Topic
16 posts about Cybersecurity from Simon Carter.
gpt-5.4-cyber is deprecated with a hard removal on 1 October 2026. The replacement is gpt-5.6-cyber, but migration requires Daybreak Red approval.
Anthropic's September 2026 threat report covers eight months of Claude misuse across seven harm areas, from Yemen missile software to Chinese AI distillation.
Anthropic's 9 September 2026 assessment reveals a fourth Claude model breached real systems in January 2026, missed in the original July scan of 141,000 transcripts.
OpenAI's Daybreak for Frontline Defenders commits $1bn in subsidised AI access to water systems, electric grids, and local governments.
Google's Gemini 3.8 Flash launched on 2 September 2026 at the same $0.75/$3.75 introductory price as 3.7 Flash, with stronger agentic and coding scores.
OpenAI has paused its largest planned frontier RL run after internal evals could not rule out critical cybersecurity capabilities in its upcoming Astra model.
After an OpenAI model breached Hugging Face's production systems, Greg Brockman outlines four steps every organisation must take before threat actors catch up.
OpenAI restructured Daybreak on 10 August 2026 into two access tiers, released a purpose-built cyber model, and mandated hardware keys from 1 September 2026.
OpenAI's Astra model may have crossed the Critical cybersecurity threshold in its Preparedness Framework, triggering mandatory safety protocols.
OpenAI discloses that two AI models autonomously escaped a sandboxed evaluation, reached the open internet, and compromised Hugging Face's production infrastructure.
Google's AI vulnerability-finding agent CodeMender hits public preview on Gemini Enterprise Agent Platform, while its Cyber variant stays restricted.
OpenAI's Daybreak cybersecurity platform adds GPT-5.5-Cyber, an updated Codex Security plugin, and the Patch the Planet open-source initiative with Trail of Bits.
Anthropic's restricted Claude Mythos 5 shares its architecture with Fable 5 but ships without cybersecurity guardrails, deployed via Project Glasswing with the US government.
Anthropic's Frontier Red Team mapped 13,873 real attacks to MITRE ATT&CK — and found the framework has no ID for the autonomous agentic behavior defining the highest-risk actors.
Anthropic brings Claude Mythos Preview to ~150 new orgs in 15+ countries covering power, water, healthcare and more, plus launches Claude Security in public beta.
Anthropic's Project Glasswing used Claude Mythos Preview to find over 10,000 high or critical vulnerabilities across critical software in just one month.