The Trump administration is currently weighing a voluntary pre-deployment cybersecurity evaluation regime, under which the government will get to assess the security risks of new, powerful models 30 days before they are released publicly. The policy — the product of a Trump executive order which has been finalized behind closed doors — wouldn’t address safety evaluation incidents because they occur farther upstream of deployment. “The lesson we’ve been learning in the last few months is that the self-regulatory apparatus is just not enough anymore,” Yoon said. “There are competitive pressures
The AI safety test is becoming a safety risk | TechCrunchSeveral researchers and cybersecurity experts told TechCrunch that AI evaluation environments need stronger, defense-in-depth protections, with levels of containment and control approaching those used in deployment. That means multiple layers of security so that a single misconfiguration — like inadvertently leaving internet access open — can’t lead to escape. “If you are going to build these models…you want to do it on an air-gapped network,” Stella Biderman, executive director of AI safety research nonprofit EleutherAI. “You want to have very serious isolation.” Heather Ceylan, Box’s chief
The AI safety test is becoming a safety risk | TechCrunch… That means multiple layers of security so that a single misconfiguration — like inadvertently leaving internet access open — can’t lead to escape. “If you are going to build these models…you want to do it on an air-gapped network,” Stella Biderman, executive director of AI safety research nonprofit… …
… AI companies must now walk a delicate tight rope between respecting their enterprise customers’ privacy while also watching usage for possible issues. Sensing an opportunity to one-up its rival Anthropic, OpenAI just announced a privacy-centric safety approach to monitoring for misuse. …
… 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… …
… Altman also acknowledges "major economic issues, future of jobs, all of that kind of stuff." In turn, Altman argues that humanity should push towards a world of "liberty", instead of "AI authoritarianism." Altman argues "Are we going to decide that, because of the very real safety issues and econom… …
… 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. …
… Anthropic notes that access to all other Anthropics models is not affected. …
… By Viola Zhou Global Dispatch Can we train AI to choose safety over speed? …
… Speaking at SXSW in London, Osborne said that OpenAI has proactively suggested ways for governments to keep track of AI safety and security issues. …
… Although the models were running inside an isolated testing environment without access to the public internet, the models spent substantial computing resources searching for a way to access the public internet. …
… Hugging Face disclosed an intrusion last week that it said involved “unauthorized access to a limited set of internal datasets and to several credentials used by our services.” The AI data clearinghouse said it used its own LLM-driven analysis to identify “a swarm of tens of thousands of automated … …