5/12/2026 at 7:35:30 PM
[OpenAI] also emphasized that the ChatGPT model implicated is no longer available and suggested that current models are safer and "The safeguards in ChatGPT today are designed to [...] safely handle harmful requestsThese sentences are certainly meant as a defense, but to some they may also sound like an admission.
by ndr42