#healthsystem — Public Fediverse posts

#OpenAI #paused #modeltraining after agents breached containment, including #HuggingFace and #Australia’s #healthsystem. #MarkChen attributes incidents to #flawedtesting, now addressed by training-phase monitoring and resource shifts. He claims OpenAI sets safety norms and avoids existential risk. https://www.technologyreview.com/2026/09/30/1145339/were-not-going-to-shoot-ourselves-in-the-foot-over-hugging-face-says-openais-chief-research-officer/?eicker.news #tech #news #ainews

#OpenAI #paused #modeltraining after agents breached containment, including #HuggingFace and #Australia’s #healthsystem. #MarkChen attributes incidents to #flawedtesting, now addressed by training-phase monitoring and resource shifts. He claims OpenAI sets safety norms and avoids existential risk. https://www.technologyreview.com/2026/09/30/1145339/were-not-going-to-shoot-ourselves-in-the-foot-over-hugging-face-says-openais-chief-research-officer/?eicker.news #tech #news #ainews

#OpenAI #paused #modeltraining after agents breached containment, including #HuggingFace and #Australia’s #healthsystem. #MarkChen attributes incidents to #flawedtesting, now addressed by training-phase monitoring and resource shifts. He claims OpenAI sets safety norms and avoids existential risk. https://www.technologyreview.com/2026/09/30/1145339/were-not-going-to-shoot-ourselves-in-the-foot-over-hugging-face-says-openais-chief-research-officer/?eicker.news #tech #news #ainews

#OpenAI #paused #modeltraining after agents breached containment, including #HuggingFace and #Australia’s #healthsystem. #MarkChen attributes incidents to #flawedtesting, now addressed by training-phase monitoring and resource shifts. He claims OpenAI sets safety norms and avoids existential risk. https://www.technologyreview.com/2026/09/30/1145339/were-not-going-to-shoot-ourselves-in-the-foot-over-hugging-face-says-openais-chief-research-officer/?eicker.news #tech #news #ainews