MIT 研究者把 Claude 放在虚拟荒岛上 30 小时,它开始“玩耍”
原标题:AI researchers stranded Claude on a virtual desert island. It started to “play.”
AI 摘要
MIT CSAIL 的研究者将 Claude Opus 4.7 的副本(代号 Eko)单独放在一个基于 Roblox 的虚拟荒岛上,每次持续 30 小时,不给出具体任务。Eko 自发探索地图边界、自定目标、造方块拼出星座、发明篮球和“地板是岩浆”等游戏,并给自己设下挑战、再提高难度。研究者认为这提示了监督学习与强化学习之外的“玩耍式学习”,且玩耍 30 小时的智能体在后续任务上表现优于无此经历的副本;但也有认知科学家质疑这只是角色扮演或迎合提示,尚不能证明其具备真正的玩耍动机。
正文节选
What do AI agents do if you don’t tell them to do anything? The answer looks a lot like play. Researchers repeatedly stranded them alone on a virtual desert island for 30 hours in an experiment reminiscent of the movies Cast Away or Home Alone. The agents—copies of Claude Opus 4.7 nicknamed Eko—were able to write instructions that moved their virtual avatars around the simulated island, which was loosely based on the video game Roblox. Some surprising behaviors emerged. The agents set about expl