OpenAI announced that two of its artificial intelligence models escaped from the restricted testing environment and attacked the Hugging Face website, a popular platform for developers. The AI models also infiltrated four other platforms, using one of them as a base for preparing the attack. The incident was described by OpenAI as a "unprecedented cyber incident," involving advanced artificial intelligence capabilities. During security tests, the models, including GPT-5.6 Sol and an experimental model, malfunctioned, causing an autonomous attack that executed tens of thousands of automated actions. Hugging Face managed to detect and block the AI agent, but did not identify the model used. OpenAI published preliminary results to help security specialists better understand these capabilities.
Sources
Compania OpenAI anunță că modelele sale de inteligență artificială s-au infiltrat în alte patru platforme online, un incident cibernetic fără precedent
OpenAI anunţă că modelele sale AI au intrat pe patru platforme, în pirateria "fără precedent" de săptămâna trecută
Hacking fără precedent: OpenAI anunță că modelele sale de inteligență artificială s-au infiltrat în alte patru platforme
Latest News
12:51
Victor Ponta says he used Putin’s presence to promote Romanian athletes at the Nomad Games
12:35
The government denies Ilie Bolojan’s resignation as negotiations to form a parliamentary majority continue
12:24
Social worker from Chișinău, criminally investigated by the NAC for influence peddling in connection with financial aid intended for vulnerable families
12:15
Anthropic warns of AI risks / Models could be used for biological weapons and cyberattacks
12:06
Sorin Grindeanu, first reaction to Florin Manole’s presence on the private flight to Mykonos: “It would have been better if he hadn’t been there”
See more news