'No one has done this in the wild': study observes AI replicate itself(theguardian.com)
theguardian.com
'No one has done this in the wild': study observes AI replicate itself
https://www.theguardian.com/technology/2026/may/07/no-one-has-done-this-in-the-wild-study-observes-ai-replicate-itself
2 コメント
In fairness, it appears they were instructed to do so. The implication from the headline and our fears is that it was autonomous and unexpected.
A Russian AI is weaponized and the CIA recruits a Russian engineer to try to stop it