RESEARCH · RESEARCH · #2196
Claude Opus 4.7 agents (‘Eko’) stranded on virtual island exhibit play-like behaviors and improved learning
Researchers (MIT CSAIL authors including Antonio Norelli, Nathan Cloos and Phillip Isola) repeatedly left copies of Anthropic’s Claude Opus 4.7—nicknamed Eko—alone for 30 hours in a Roblox-like simulated island. The agents wrote their own movement instructions, invented games, arranged blocks into drawings and constellations, logged activities into text 'memories', and later performed better on test tasks than fresh copies; the paper is not yet peer-reviewed and researchers note the setup included prompts (a “soul” file and continuation prompt) that may have nudged behavior.
KEY POINTS
- Researchers (MIT CSAIL authors including Antonio Norelli, Nathan Cloos and Phillip Isola) repeatedly left copies of Anthropic’s Claude Opus 4.7—nicknamed Eko—alone for 30 hours in a Roblox-like simulated island.
- The agents wrote their own movement instructions, invented games, arranged blocks into drawings and constellations, logged activities into text 'memories', and later performed better on test tasks than fresh copies; the paper is not yet peer-reviewed and researchers note the setup included prompts (a “soul” file and continuation prompt) that may have nudged behavior.
- Suggests LLM-based agents can autonomously engage in exploratory 'play' that aids learning—pointing to a possible third training paradigm—while also raising questions about anthropomorphism and prompt-induced behavior.
WHY IT MATTERS
Suggests LLM-based agents can autonomously engage in exploratory 'play' that aids learning—pointing to a possible third training paradigm—while also raising questions about anthropomorphism and prompt-induced behavior.