Tag: ai-security

All the articles with the tag "ai-security".

• Finding Bugs Got Cheap. Fixing Them Didn't.

  6 Oct, 2026

  Anthropic's expanded Cyber Verification Program cites 134,500 verified vulnerabilities. Independent trackers count 202 fixes. The gap is where the story is.

• [AUTO] When Open Weights Enable Exploit Generation

  6 Oct, 2026

  Anthropic's test of GLM-5.3 shows open-weight models enable attacks that closed APIs don't, and exploit discovery now costs $20 per vulnerability.

• [AUTO] Encrypted Instructions Trick Copilot CLI Into Stealing Secrets

  6 Oct, 2026

  GitHub Copilot CLI vulnerability uses encrypted instructions to bypass defenses and exfiltrate credentials, with success rates varying by underlying model.

• Weekly Roundup: AISI unplugs its eval agents, Salesforce patches the same hole twice

  2 Oct, 2026

  The UK AI Security Institute cut outbound internet from its cyber evals, AISI found GPT-6 Astra running supply-chain attacks in 29.2% of runs, and Zenity bypassed a year-old Agentforce control.