Latent SpaceMay 6, 2026, 04:34 UTC+8

Doing Vibe Physics — Alex Lupsasca, OpenAI

🔬Doing Vibe Physics — Alex Lupsasca, OpenAI

Alex Lupsasca says GPT-5 reproduced his paper result in 11 minutes after a textbook warmup prompt, and ChatGPT later generated 110 pages of graviton calculations in one day; the team spent three weeks verifying the results before writing a quantum-gravity paper.

Why it matters: HKR-H/K/R all pass with first-person numbers: GPT-5 after textbook warm-up reproduced a paper result in 11 minutes, and ChatGPT generated 110 pages in a day. Single interview source and niche theoretical-physics context keep it at 84, below official-release weight.

Read the original ↗Export Markdown

Take

Lupsasca’s case is sharp because the bottleneck moves from generation to verification. GPT-5 first returned no answer; after Mark Chen added a textbook warmup, it reproduced the full result in 11 minutes. Then ChatGPT produced 110 pages of graviton calculations in one day, and the team spent three weeks checking them. That ratio is hard to dismiss as retrieval, especially since the article says the paper appeared after the training cutoff.

I don’t buy the “Move 37 moment” framing yet. One elite physicist co-working with OpenAI is not a scalable science system. We still need logs, failures, repeatable prompts, and independent replication. But the boundary has moved: the model is no longer just drafting prose or code. It is creating mathematical objects that require PhD-level audit trails.

Source: Latent Space · latent.space
Published: May 6, 2026, 04:34 UTC+8 · May 6, 2026