Experts weigh limits of current AI safety testing techniques
1 outlet · 1L · 0C · 0R
First reported
1 / 1
We're still gathering coverage of this story.
A neutral account and the side-by-side comparison are written once at least 2 outlets have reported it. Here is the coverage so far.
Left1 article
Centre0 articles
No coverage yet.
Right0 articles
No coverage yet.
- 6 Oct 15:30 First Scientific AmericanLC What makes a good AI safety test? Experts explain why even the best techniques may not be powerful enough
Times are when each article was published, or when we first saw it if the outlet gave no time.