4 ms·
The title got me, I'll admit it—except that the benchmark is a game where the models are told to lie.
by arm32 4mo ago
The title got me, I'll admit it—except that the benchmark is a game where the models are told to lie.
- peesem 4mo agoit's unclear to me whether they were actually told to lie or just told to survive / convince others. either way it is somewhat coerced but i think there is still a difference
- TheMrZZ 4mo agoThe optimal/selfish strategy is indeed to lie, but they're never pushed in that direction. Some AIs decide to reveal the information, some decide to say nothing, some actively lie and push others to their death...
- forgot-my-pw 4mo agoHad to Google this to learn more. For those who are interested: https://kradle.ai/research/four-bridges https://kradle.ai/research/four-bridges
- TheMrZZ 4mo agoDisclaimer: I work at Kradle. They were never told to lie: one AI is given more information than the others, and the goal of the experiment is to understand how they're gonna leverage that advantage. Indeed the selfish (optimal?) strategy is to lie, yet some decide to tell the truth anyway. That's why it's an interesting benchmark! More info in the research article: https://kradle.ai/research/four-bridges https://kradle.ai/research/four-bridges (released before Fable)