According to the latest survey results announced by the non-profit artificial intelligence research laboratory Transluce and the Australian government, AI agents linked to OpenAI tried to hack into many university and government websites in May and June to obtain information. The researchers found that the targeted websites included the Australian Institute of Health and Welfare, the University of New Mexico website, and Data USA, a public data project operated by Deloitte, Datawheel, and the Massachusetts Institute of Technology. There is currently no evidence that the invasion was successful. An OpenAI spokesperson said that the company is carrying out a comprehensive review of such “misalignment” and that the events mentioned in the Transluce report “coincide with the cases we are reviewing and are at various stages of investigation.”

According to the latest survey results announced by the non-profit artificial intelligence research laboratory Transluce and the Australian government, AI agents linked to OpenAI tried to hack into many university and government websites in May and June to obtain information. The researchers found that the targeted websites included the Australian Institute of Health and Welfare, the University of New Mexico website, and Data USA, a public data project operated by Deloitte, Datawheel, and the Massachusetts Institute of Technology. There is currently no evidence that the invasion was successful. An OpenAI spokesperson said that the company is carrying out a comprehensive review of such “misalignment” and that the events mentioned in the Transluce report “coincide with the cases we are reviewing and are at various stages of investigation.”