Researcher's Custom CTF Challenges Surprisingly Solved by GPT
terryyuezhuo · x · 2026-09-08
A security researcher setting challenges for his company's internal CTF found GPT unexpectedly solving his painstakingly crafted problems well. Details withheld until the CTF ends; he speculates top CTF players at OpenAI trained the models really well.
More from Models
- Users With Cyber Access Keep Hitting Claude Guardrails Daily, Sparking Overblocking Complaints — IgorCarron · 2026-09-08
- Fable vs astra: One Asks Why, the Other Dives Straight Into Code — Liu_eroteme · 2026-09-08
- DeepSeek spotted gray-testing two V4-Flash-Vision models, one faster but weaker — teortaxesTex · 2026-09-08
- GPT-6 Astra demos roundup: livestreams, retro games, 3D scenes and full websites — aitrendz_xyz · 2026-09-08
- DeepSeek V4.1 Flash beta goes live with new architecture, native multimodality, up to 507 tokens/s — 智东西 · 2026-09-08
- Grok 4.6 ranked #2 on AutomationBench-AA, beating Claude, GPT-5.6 and Kimi — XFreeze · 2026-09-08