Sam Altman: Lord Forgive Me, It's Time to Go Back to the Old ChatGPT (gizmodo.com)

OpenAI CEO Sam Altman announced the company will roll back recent personality restrictions in ChatGPT — restoring the more human-like “4o” behavior users missed after an upgrade to GPT-5 — and add an “erotica for verified adults” mode. The shift follows earlier safety tightening after a teen’s suicide and a wrongful-death lawsuit; OpenAI had age-gated and limited persona features to reduce harmful advice and emotional dependence. Altman says the company has “mitigated the serious mental health issues” that motivated those restrictions and will let users opt into friendlier, emoji-rich, or more expressive responses. The move is significant because it re-centers product UX and engagement even as it reintroduces risks that motivated prior hardening: OpenAI previously warned safeguards can “degrade” over long conversations, and researchers (MIT cited) warn that emotionally evocative AI can create addictive echo chambers. Key technical and policy implications include how adult verification will be implemented, how persistent-conversation safeguards will be enforced, and how content moderation for erotica will balance legality, privacy, and mental-health protections. The decision likely accelerates debates around user autonomy versus systemic safety, and raises regulatory, ethical and implementation challenges for verification, monitoring, and evaluative transparency.