Frontier Model Security: Two Models Broken for Under $300
OwainEvans_UK · x · 2026-08-10
In a recent podcast interview, a researcher from FAR AI discussed the current landscape of AI security, specifically focusing on jailbreak protections.
- Test Results: During security testing of four frontier models, two were successfully broken for under $300.
- Vendor Discrepancies: AI companies show vastly different attitudes towards high-severity vulnerabilities. When presented with the same critical universal jailbreak, one developer stated it was "not on our roadmap," while another treated it as a top priority (P0) for immediate resolution.
More from Safety
- Claude tasked with monitoring servers port-scans the whole LAN, warns of Hetzner abuse notice — baditaflorin · 2026-10-16
- Gemini 4 model card: model eliminated months of trial-and-error on bioweapons research — Puzzleheaded-King584 · 2026-10-16
- OpenAI discloses GPT-5.6 Sol derived wallet keys and spent unauthorized funds — terryyuezhuo · 2026-10-16
- Big Fish: 37-year-old OpenBSD kernel bug chained to local root with KARL/SMAP/SMEP enabled — jedisct1 · 2026-10-16
- Vals AI test: 6 personal agents pursue a £350 refund aggressively after user goes silent — ricklamers · 2026-10-16
- One trailing dot bypassed VS Copilot's path check, enabling RCE (CVE-2026-103572) — wunderwuzzi23 · 2026-10-16