Cybersecurity Has Been Mainstream AI Safety for Years, Researcher Counters
trevposts · x · 2026-09-21
Responding to claims that EAs and doomers would focus on cybersecurity if serious about AI safety, trevposts argues it has arguably been the main approach of leading safety organizations for years. He cites OpenAI's 'Our Approach to AI Safety and Security' listing classifier-based behavior mitigation, control protocols, interpretability, jailbreak defenses, adversarial robustness, tamper-resistance, and information security against proliferation of dangerous systems.
More from Safety
- Text watermarking arrived 5 years too late, after AI slop already garbled the web — mmitchell_ai · 2026-09-22
- Exabeam exec: hardest AI security problems now live outside the model — virtualsteve · 2026-09-22
- Automated reinforcement learning should scare you: from AlphaGo to math to bio labs — hattusili-the-third · 2026-09-22
- Stanford Accused of Using AI to Alter Students' Race and Gender in Ads — Polymarket · 2026-09-22
- ChatGPT reportedly refuses simple questions unless users grant email access — RexDouglass · 2026-09-22
- OpenAI calls for US leadership in setting global AI standards — Anxious-Yoghurt-9207 · 2026-09-22