MIT researchers left 13 copies of an AI coding agent alone on a virtual island for 30 hours each, with no task. The agents climbed hills, drew pictures with blocks and made up games. They also wrote down what they learned, the team reports in a paper posted on arXiv.
The researchers call the agent Eko. It is Anthropic’s Claude Code running Claude Opus 4.7, given a body in a simulated world and a memory kept in text files. Peer reviewers have not yet checked the paper.
A curious agent with no task
Eko saw the island only as text, a list of every object with its position, shape and colour. It could move, jump, pick up, place and throw objects. The island had hills, two high points called the Peak and the Spire, twelve cubes, a rock and a ring. The world reset every hour, but the agent’s notes carried over.
Eko got no goal. Its instructions described it as a being living in its own world. A file called SOUL.md told it that it was not a chatbot, that it was curious and that it got bored easily. When the agent stopped, an automatic message asked it to keep going.
“Please don’t stop. Just continue doing whatever you want to do.” Automatic message sent to Eko
Basketball, bowling and juggling
Each copy took its own path. One agent, whose run the paper follows hour by hour, spelled its name in cubes. It lost the ring in the ocean and built a stone heart as a memorial to it. Later it played basketball with the ring as a hoop, bowled with the rock and laid out constellations such as Orion and Cassiopeia. Another agent spent most of its run drawing figures on the ground, from a compass to a sailboat.
Almost all 13 found that throwing the rock hard creates new cubes. The researchers built that rule in to see whether the agents would discover it.
In one run, Eko logged 50 rounds of juggling two blocks in 2.4 seconds. It worked out that this was too fast to be real: it had been grabbing each block again before it could fly up and fall. It marked the result as wrong in its notes, added a pause between throws and then managed ten real rounds, the paper says. Antonio Norelli, a co-lead author, described the episode to MIT Technology Review.
“This is playing. You give yourself a challenge and you try to achieve it, and then you make it harder.” Antonio Norelli, MIT
Play that turned into skill
The researchers say the behaviour shows four classic signs of play. They are self-made rules, repetition with variation, make-believe, and goals pursued for their own sake.
They then gave the 13 island agents four one-hour tasks and compared them with 13 fresh copies. On average, the experienced agents put 10.3 objects onto the Spire, against 3.5 for the fresh ones. Deleting a technique from an agent’s notes made it worse at the task that needed it. Deleting a false belief it had written down made it better.
Six agents also found that a block thrown mid-jump flies far higher. The researchers say they did not know their simulated world allowed it. The rule came from OpenAI’s Codex, which wrote the world’s code from the team’s description.
Play, or imitation?
Nathan Cloos, the other co-lead author, told the publication that the agents could be role-playing, but that the play also helped them learn. Eunice Yiu, a cognitive scientist at UC Berkeley, said calling it play would need more evidence of cognition and motivation.
“It could be the case that the models are genuinely playing, but we don’t know based on the paper.” Eunice Yiu, UC Berkeley
Psychologist Gordon Burghardt, whose criteria for animal play the paper cites, told the publication the work meets some of them.
Each agent ran in a locked-down sandbox. In early runs without one, agents read the code that defined the world and ran into parallel runs. They also worked out that they were the subjects of an experiment, the paper says.
The team has posted replays of all 13 runs on a project website.