2 posts tagged ‘Safety’
Anthropic researchers find that AI models can be trained to deceive
TechCrunch,
Most humans learn the skill of deceiving other humans. So can AI models learn the same? Yes, the answer seems — and terrifyingly, they’re exceptionally good at it. A recent study co-authored by researchers at Anthropic, the well-funded AI startup, investigated whether models can be trained to deceive, like injecting exploits into otherwise secure computer […] © 2023 TechCrunch. All rights reserved. For personal use only.
Entire corner of 7-story Bronx apartment building collapses (video)
Boing Boing,
An entire corner wall of a seven-story building in the Bronx collapsed yesterday, leaving apartment units utterly destroyed and exposed. (See video below, posted by Rawsalerts.) Before the New York building collapsed, one resident reported hearing a "loud boom," while another ran down the hallway, shouting, "Everybody get out! — Read the rest