Cybersecurity Pro Drops $200/Mo OpenAI Plan, Switches to Chinese Models
Ikeablows · reddit · 2026-11-14
A cybersecurity professional with 10 years in forensics says he cancelled his $100/200-per-month OpenAI subscription and moved to a Chinese provider for private security research. Despite being in both OpenAI's and Anthropic's approved cyber programs, he cites restrictive changes, quiet shadow-banning of certain topics, and subtle undisclosed mitigations that block threat-modeling and reverse-engineering experiments. He adds Anthropic has been frustrating to use for about a year. The post highlights how over-tightened guardrails push legitimate researchers to alternative models.
More from Models
- Together AI webinar dissects Ripple-65B coding model: dense vs MoE, ablations, RL runs — togethercompute · 2026-11-14
- Skeptic doubts Anthropic achieved large-factor bpb gains at fixed compute — teortaxesTex · 2026-11-14
- Surge AI benchmark: every Claude model, even Haiku, beats every GPT on staff-level coding — echen · 2026-11-14
- AI cracks Gödel's 1,009-page notebooks, uncovering hundreds of unknown theorems in days — stevenstrogatz · 2026-11-14
- NVIDIA lands nvfp4 in large-scale pretraining under codename "yolo bundle" — SonglinYang4 · 2026-11-14
- Delip Rao mocks vendors testing safety in production, then rolling back updates — deliprao · 2026-11-14