AI researchers stranded Claude on a virtual desert island. It started to play.
News Source : MIT Technology Review
News Summary
- Researchers stranded AI agents alone on a virtual desert island for 30 hours.
- The agents set about exploring the limits of the world and came up with their own goals.
- The researchers say Eko’s behavior has what classic studies, including work by the influential University of Tennessee psychologist Gordon Burghardt, suggest are four hallmarks associated with play in humans and animals.
- But was Eko actually playing, or was the word-prediction engine simply role-playing what a playful agent would do in this scenario?
AI systems are usually trained through supervised learning, where they copy examples provided by humans, or through reinforcement learning, where they pursue a reward through trial and error.
Never miss a story from us, subscribe to our newsletter