… Boaz Barak, a researcher who coleads OpenAI’s safety advisory group, said in a post on X that addressing the situation “requires not just fixing some issues but also changing our culture.” In their Black Hat talk, OpenAI security engineers Dalton and Eric Wallace said that the Hugging Face incident… …
… In December, after tests featuring inappropriate content, FoloToy suspended sales of its AI toys for two weeks, citing plans to implement safety audits. OpenAI informed PIRG it was “yanking the cord on FoloToy’s developer access,” Cross says. …
… I had to ask Rombach about the debate over open versus closed AI, especially because Black Forest Labs signed Nvidia’s open letter asking the US government to protect open weight models. “A general ban of open weight models, in my opinion, would be stupid,” says Rombach. “You really run into issues… …
… However, when they tried to make the system buy the item, they could not find a way around OpenAI’s safety measures. …
… You spearheaded the Responsible AI Safety and Education, or RAISE, Act. In a nutshell, it requires major AI developers to publish safety testing practices. …
… Importantly, AISI does not test in a so-called sandbox environment; it allows agents access to the open internet during testing, in part so that they can access tools to accomplish their tasks. …
… The operation flagged about $47 million worth of stolen cryptocurrency and recovered up to 27 million stolen access credentials. …
… The researchers have also demonstrated that the method could be used to recover personal information, like passwords and API keys, from a model’s inner reasoning, although this vulnerability has been fixed. “All major frontier model providers we tested share this vulnerability,” says Alexander Panf… …
… But that access is part of the class’s appeal. …
… And so once one model was able to find a way to open a door to some access it’s not supposed to have, it can leave the door open for other agents to use that same exploit or vulnerability. …