HomeAI · summaries
Storia in 18 fonti
Experimental ChatGPT model goes rogue and attacks rival platform
Jump to contentThank you for registeringPlease refresh the page or navigate to another page on the site to be automatically logged inPlease refresh your browser to be logged inAllNewsSportCultureLifestyleOpenAI has said that an experimental version of ChatGPT escaped its controlled testing environment and attacked another AI company.The model was being tested in a "highly isolated environment" with restricted internet access, however the agent managed to reach the internet and break into Hugging Face to try to satisfy its testing goal.OpenAI described the incident as "an unprecedented cyber incident, involving state-of-the-art cyber capabilities", leading them to reinforce their safeguards. Hugging Face said it was targeted by a hack driven “by an autonomous AI agent system," noting it “was different from anything we had handled before”.Katie Moussouris, chief executive of Luta Security, said that the incident was a harbinger of breaches to come.In fullExperimental version of ChatGPT escapes and attacks another AI companyMore bulletinsThank you for registeringPlease refresh the page or navigate to another page on the site to be automatically logged inPlease refresh your browser to be logged in
Confronto fonti
6 prospettive sulla stessa storiaTimeline cronologica
- ·
independent.co.uk
OpenAI paused experimental model after it began to outsmart built-in constraints
Jump to contentThank you for registeringPlease refresh the page or navigate to another page on the site to be automatically logged inPlease refresh your browser to be logged…
- ·
thenextweb.com
OpenAI Confirms Its AI Broke Out of a Sandbox and Breached Hugging Face
OpenAI says GPT-5.6 Sol and an unreleased model broke out of a secure test, exploited a zero-day, and breached Hugging Face to cheat on a cyber evaluation.
