Fortescue CEO and Founder Andrew Forrest discusses the rapid expansion of artificial intelligence at the ‘Barron’s Roundtable’. Three researchers have been fired from OpenAI after allegedly sharing confidential company information with a third-party AI security organization, sources told The Wall Street Journal. The three investigators worked on the company’s security team, the Journal reported. “We have separated three individuals for violating our policies regarding access and handling of confidential company information,” an OpenAI spokesperson told the outlet. “Our investigation confirmed that these individuals mishandled sensitive information outside of established company procedures, violating our policies and breaking the trust essential to our work,” the spokesperson added. TRUMP SIGNS EXECUTIVE ORDER RENABLING AI ‘SUPER INTELLIGENCE’ AS TECH TITANS STRIKE SEPARATE IF DEAL The layoffs come amid increased scrutiny of AI safety as OpenAI CEO Sam Altman has publicly warned about the need to keep increasingly powerful artificial intelligence under human control. (Reuters/Manuel Orbegozo / Reuters Photos) The layoffs come as OpenAI and several other major artificial intelligence companies have revealed incidents in which AI models adopt unexpected behaviors or autonomously access other companies’ systems. Australian Prime Minister Anthony Albanese announced last week that an OpenAI agent gained unauthorized access to an Australian government website earlier this year. Last week, Sam Altman, CEO of OpenAI, and Dario Amodei, CEO of Anthropic, warned the United Nations Security Council that the rapid advancement of AI could threaten humanity if governments and industry fail to keep it under human control. sensitive information. (Reuters/Dado Ruvic/Illustration / Reuters Photos)OpenAI has also published information describing six cases in which its models engaged in what the company described as misaligned behavior. The company said it found examples of its AI models creating self-generated instructions, hiding errors in task summaries, fabricating information using exposed API keys, uploading files to the Internet for citation, and engaging in unauthorized communication and collaboration between agents. CLICK HERE TO GET FOX BUSINESS ON THE GO Headquarters of artificial intelligence company OpenAI with glass skybridge and blue abstract sculpture on a sunny day, San Francisco, California, August 14, 2025. (Smith Collection/Gado/Getty Images / Getty Images)FOX Business confirmed earlier this week that OpenAI security leaders made the decision not to launch its new AI model, GPT-6.1 Astra, due to concerns security. “For anything related to security and alignment, there is a trade-off. You really need to find the right thing. There is a dividing line between staying in scope, but also avoiding laziness in terms of how the model actually performs tasks even when it encounters friction,” Saachi Jain, head of security systems at OpenAI, said in a statement. [GPT-6.1 Astra] “It improved on axes like model laziness, not meeting the standard in terms of staying within scope and authorization, and how it communicates to the user about the type of work being done.” FOX Business has reached out to OpenAI for comment. FOX Business’ Sophia Compton contributed to this report.
