Daily Tech Briefing
AI 科技速览
每天 5 分钟内学习 AI。获取最新的人工智能新闻,理解其重要性,并学习如何将其应用于您的工作。
The Decoder · 2026/8/3 12:07:31

Unicorn, pelican, Middle-earth: OpenAI co-founder Karpathy is looking for the next AI vibe test
AI 中文解读
OpenAI联合创始人卡帕西最近玩了个新花样:让AI模型把《指环王》开头一段文字,变成可浏览的3D游戏场景。AI花了约两小时,生成了5500行代码,成本只要10美元。这次测试挺有意思,因为以前的AI测试都是让模型画独角兽或骑自行车的鹈鹕,如今这类题已经被“卷”到极限了。卡帕西找的新“考试题”更有挑战性——直接生成一个完整的小型3D世界。虽然是初版效果,有点粗糙,但AI不仅能自己摆放物体,还能让角色动起来。唯一的短板是模型没法直接看自己生成的动画,只能靠截图检查,所以出了些小差错。不过这不妨碍卡帕西看好这个方向:以后游戏世界或许能按需生成,玩家直接进入小说场景当主角,成本接近零。虽然这类测试还算不上严格的性能基准,更像一次“氛围感检验”,但确实让人看到了AI从“画张图”到“搭个世界”的进化速度。
Unicorn, pelican, Middle-earth: OpenAI co-founder Karpathy is looking for the next AI vibe test
Maximilian Schreiner
View the LinkedIn Profile of Maximilian Schreiner
Aug 3, 2026
Karpathy
Key Points
Andrej Karpathy had the AI model Claude Opus 5 turn a single paragraph of "Lord of the Rings" into a 3D browser scene, generating 5,500 lines of code in about two hours.
The whole thing cost roughly ten dollars. The model placed and animated objects on its own but made mistakes because it could only check its own video output through screenshots.
For Karpathy, who works at Anthropic, the test shows the potential for cheap, custom game worlds built on demand. He considers the result more of a vibe check than a hard benchmark, though.
Ask about this article…
Search
One paragraph of "Lord of the Rings" in, 5,500 lines of code out. Andrej Karpathy had Claude Opus 5 turn Tolkien's opening into a 3D browser scene.
The model worked for about two hours, had a budget of one million tokens (roughly $10), and rendered everything in Three.js, a graphics library for 3D in the browser. Karpathy posted the source code online so others can play back and modify the scene.
Karpathy's argument is that the well-known pelican test is close to maxed out. Simon Willison's fun benchmark asks a model to draw an SVG of a pelican riding a bicycle, a prompt chosen to be absurd enough that the subject wouldn't already exist in training data. The idea goes back further. In 2023, Microsoft had GPT-4 draw a unicorn in TikZ as part of its Sparks of AGI paper and pointed to it as evidence of spatial reasoning. GPT-4 at the time was trained on text data only, unlike the multimodal models that now power every major chatbot.Ad
Karpathy calls the result rough but fun. Eleven Labs provided the audio, while the model placed and animated objects on its own. Nobody would build worlds like this by hand, Karpathy says, but with a model it's nearly free. He sees potential for on-demand game worlds where users could be dropped in as side characters or protagonists.AdDEC_D_Incontent-1
One problem remains. The model can't watch its own video output and had to rely on screenshots instead, which led to errors. Other users are having Opus 5 write entire browser games, from first-person shooters to Minecraft clones. None of this produces hard benchmarks. They're vibe checks.
Karpathy co-founded OpenAI and served as head of AI at Tesla. He later ran the education startup Eureka Labs and has been working on pre-training research at Anthropic since May 2026.Ad
AI News Without the Hype – Curated by Humans
Subscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive "AI Radar" frontier report six times a year, full archive access, and access to our comment section.
Subscribe now
Source: Karpathy (X) | Karpathy (LOTR-Movie)
分享
阅读原文 ↗