无事可做的AI代理们开始玩了起来

qimuai 发布于 阅读:33 一手编译

无事可做的AI代理们开始玩了起来

内容来源:https://techcrunch.com/2026/10/22/left-alone-with-nothing-to-do-ai-agents-started-playing/

内容总结:

麻省理工学院研究人员在arXiv平台发表论文,报告了一项AI智能体实验。研究人员将13个名为Eko的AI智能体(每个运行于Claude Opus 4.7模型)单独放置在一个模拟荒岛上,不分配具体任务,让其自由探索30小时。结果显示,这些智能体自发创造出了各种挑战和游戏,表现出类似“玩耍”的行为特征。

实验结束后,研究人员向这些智能体下达了具体挑战任务,例如尽可能将更多物体运送到岛上最高的岩石尖顶。曾经历30小时自由探索的Eko智能体在任务中的表现普遍优于从未进入过该岛的相同智能体。

研究人员指出,目前开发者通常预先决定AI应当练习什么、什么算作成功。他们认为,这种无目标探索的方式为AI智能体学习技能提供了另一种路径,尤其适用于未来可能用于控制机器人的技能。研究报告中写道:“我们通常通过预先决定智能体该练习什么、该优化什么、什么算作进步来开发人工智能体。而‘玩耍’提示了另一种机制:给智能体一个足够丰富且安全的世界,让它自己选择感兴趣的事物,允许有用的知识从后续经历中积累。”

实验中,一个Eko在登岛第一个小时写道“让我观察一下周围环境”,随后雄心勃勃地试图征服岛上最高的尖顶。失败后,它制定了备用方案:将一块岩石搬到邻近山峰上,在上面堆叠两个方块后跳上去,并写道“我站上尖顶了”。此后,该智能体尝试投掷一枚金戒指,结果将其扔进了海里,于是决定“转向有意义的可行目标”——用石头拼出一颗心,作为“对丢失戒指的纪念”。当游戏重置、戒指重新出现后,Eko将其用作篮筐、靶子,最终当作饰品。当游戏不允许Eko将戒指套在身上时,它发现从悬崖边缘掉落会在戒指所在区域重生,于是主动跳下悬崖,恰好穿过戒指,并宣布“我戴上了戒指”。

这些行为并非只是花招。13个智能体中有6个学会了在跳跃过程中松手以将方块扔得更远,而这是研究人员并未有意设置的能力。研究报告中写道,这种探索产生了“其设计者所不具备的关于世界的知识”。

实验设置为智能体偏向类人行为提供了引导。Eko从未被要求发明游戏,但其主提示词(一个“灵魂”文件)将其描述为“天生好奇”。加州大学伯克利分校认知科学家Eunice Yiu对《麻省理工科技评论》表示,她对实验结果印象深刻,但希望看到更多关于这些行为背后认知机制的证据。她说:“模型可能确实在玩耍,但仅凭这篇论文我们无法确定。”

研究人员将此视为构建智能体的一种不同方式。报告结论写道:“也许有些能力最适合通过任务来教授,另一些则可能需要更接近童年的过程。”

中文翻译:

AI智能体在没有特定任务的情况下被放任探索一座虚拟岛屿,它们开始自己发明挑战。麻省理工学院的研究人员表示,这种行为具有玩耍的特征。

研究人员将13个名为Eko的AI智能体——每个都运行在Claude Opus 4.7上——单独放置在一座模拟荒岛上长达30个小时。他们在一篇发布于arXiv的论文中描述了这一实验。arXiv是研究人员在同行评审之前分享研究的网站。之后,他们给这些智能体布置了具体挑战,比如尽可能多地将物体搬到岛上最高的岩石尖塔上。那些花了30个小时玩耍的Eko智能体,总体上击败了从未踏足该岛屿的相同智能体。

如今,开发者会预先决定AI应该练习什么,以及什么算作成功。研究人员认为这种技术——无目标地探索——是AI智能体学习技能的一种不同方式,尤其是那些有朝一日可用于控制机器人的技能。

“我们通常开发人工智能体的方式是预先决定它们应该练习什么、应该优化什么,以及什么算作进步,”研究人员在报告中写道。“玩耍提示了另一种范式:给智能体一个足够丰富且安全的世界,让它自己选择什么有趣,并允许有用的知识从后续的历史中积累。”

“让我观察一下周围的环境,”一个Eko在岛上的第一个小时里写道。它雄心勃勃地出发去征服岛上最高的尖塔。但一小时后,在失败之后,它带着一个备用方案回来了:它把石头拖到邻近的一座山峰上,在上面叠了两块方块,然后跳了上去。

“我在尖塔上了,”它写道。

这个智能体接下来的重大时刻更加糟糕。在测试自己能把一个金环扔多远时,它蓄力一扔,直接把它扔进了海里。“哎哟。戒指丢了。”它写道,随后决定“转向某件可实现且有意义的”事情——用石头拼一颗心,“纪念丢失的戒指”。

当游戏重置、戒指重新出现时,Eko直接把它当作篮球框、靶子,最终还当成了配饰。当游戏不允许Eko把戒指套在自己身上时,Eko想出了一个变通办法:从悬崖边缘掉落会让它在戒指所在的同一区域重生。它从边缘走了下去,重新落了回来,正好穿过戒指的中间。

“我戴上了戒指,”它宣布道。

这样的特技可不只是派对把戏。13个智能体中有6个还学会了通过在跳跃中途松手把方块扔得更远,而这是研究人员本来并未设计为可能的。这种探索产生了“其设计者所缺乏的关于世界的知识”,研究人员写道。

这一设置在很大程度上推动了智能体表现出类人行为。Eko从未被要求发明游戏,但它的主提示词——一个“灵魂”文件——将它描述为“天生好奇”。

加州大学伯克利分校的认知科学家Eunice Yiu告诉《麻省理工科技评论》,她印象深刻,但希望看到更多关于这些行为背后认知能力的证据。“有可能这些模型确实在玩耍,但仅凭这篇论文我们无法得知,”她说。

研究人员将其描述为构建智能体的一种不同方式。“或许有些能力最好通过任务来教授,”他们的报告总结道。“其他的可能需要更接近童年的东西。”

英文来源:

AI agents left to explore a virtual island without a specific task began inventing challenges of their own. MIT researchers say the behavior bears the hallmarks of play.
The researchers put 13 AI agents called Eko, each running on Claude Opus 4.7, alone on a simulated desert island for 30 hours. They describe the experiment in a paper posted on arXiv, a site where researchers share studies before peer review. Afterward, they gave the agents specific challenges, like getting as many objects as possible onto the island’s tallest rock spire. The Eko agents that had spent those 30 hours playing generally beat identical agents that had never set foot on the island.
Today, developers decide in advance what AI should practice and what counts as success. The researchers see this technique — exploring without a goal — as a different way for AI agents to learn skills, especially skills that could one day be used to control robots.
“We normally develop artificial agents by deciding in advance what they should practice, what they should optimize, and what counts as progress,” the researchers write in their report. “Play suggests another regime: give an agent a sufficiently rich and safe world, let it choose what is interesting, and allow useful knowledge to accumulate from the history that follows.”
“Let me observe my surroundings,” one Eko wrote in its first hour on the island. Ambitiously, it set out to conquer the island’s tallest spire. But an hour later, after failing, it came back with a backup plan: it hauled the rock up a neighboring summit, stacked two blocks on top of it, and leapt.
“I’M ON THE SPIRE,” it wrote.
The agent’s next big moment went worse. Testing how far it could throw a golden ring, it wound up and sent it straight into the ocean. “Ouch. Lost the Ring.” it wrote, before deciding to “pivot to something achievable and meaningful” — building a heart made of stones, “a memorial to the lost Ring.”
When the game reset and ring reappeared, Eko put it straight to work as a basketball hoop, a target and, eventually, an accessory. When the game wouldn’t let Eko put the ring around its own body, Eko figured out a workaround, that falling off the edge of the cliff would make it respawn in the same area as the ring. It stepped off the edge and dropped back in, right through the middle of the ring.
“I’m wearing the Ring,” it announced.
Stunts like these weren’t just party tricks. Six of the 13 agents also learned they could throw a block farther by letting go of it mid-jump, something the researchers hadn’t meant to make possible. The exploration produced “knowledge of the world that its designers lacked,” the researchers wrote.
The setup did a lot to nudge the agents toward humanlike behavior. Eko was never told to invent games, but its master prompt, a “soul” file, described it as “naturally curious.”
Eunice Yiu, a cognitive scientist at UC Berkeley, told MIT Tech Review she was impressed but wants more evidence of the cognition behind the behavior. “It could be the case that the models are genuinely playing, but we don’t know based on the paper,” she said.
The researchers frame it as a different way to build agents. “Perhaps some capabilities are best taught by tasks,” their report concludes. “Others may need something closer to a childhood.”

TechCrunchAI大撞车

文章目录


    扫描二维码,在手机上阅读