Safe development of frontier AI is turning into a scarce, expensive input of its own: OpenAI has slowed training, imposed a roughly 20% compute tax to watch its strongest models, and independent investigators say they can no longer keep up with swarm-scale agent tests by hand. The argument, laid out in a Bloomberg Odd Lots newsletter by Joe Weisenthal and Tracy Alloway, is that the next constraint on the labs may not be only chips, power, or data, but the cost of running research in environments that models cannot break.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...
Contact us
Questions, corrections, or ideas? Send us a message here. We aim to reply by email within 24 hours.
Thank you. Your message has been sent. We aim to reply by email within 24 hours.Please enter your email address and a message.We could not send your message. Please try again in a little while.