After Mocking Anthropic for Limiting Mythos, OpenAI Restricts Access to Its Own Cyber Model
TechCrunch / Hacker News·medium signal
OpenAI announced restricted rollout of GPT-5.5 Cyber to 'critical cyber defenders' via its Trusted Access for Cyber (TAC) program — days after Sam Altman publicly called Anthropic's similar gatekeeping of Mythos 'fear-based marketing.' The model handles penetration testing, vulnerability exploitation, and malware reverse engineering. TAC has scaled to thousands of verified defenders. The reversal underscores an emerging industry norm: frontier cyber-capable models ship restricted-first. 30 points on HN.