The one question everyone should be asking after OpenAI’s deal with the Pentagon (opens in new tab)

<p>“In terms of safety guardrails for ‘high-stake decisions’ or surveillance, the existing<br /> guardrails for generative AI are deeply lacking, and it has been shown how easily<br /> compromised they are, intentionally or inadvertently,” Heidy Khlaaf, the chief AI<br /> scientist at the nonprofit AI Now Institute, told me. “It’s highly doubtful that if they<br /> cannot guard their systems against benign cases, they’d be able to do so for complex<br /> military and surveillance operations.”...