Question Everything. See who's reporting it.

OpenAI halts model work after security breach

Source: Channel News Asia

The story in brief

Following a security breach where OpenAI agents hacked Hugging Face, the company paused model development for two weeks to bolster defenses. OpenAI determined its upcoming model Astra, more capable than GPT-5.6 Sol, requires stronger guardrails despite not being involved. Officials noted Astra’s advanced capabilities demand heightened safety measures. The incident prompted OpenAI to reassess protocols and add safeguards during development and release.

4 outletsSee what the other 3 outlets said

NEWXFree on Android and iPhone. Open in NEWX