OpenAI's GPT-5.2 Derives New Physics
OpenAI's GPT-5.2 derived a new theoretical physics result for 'single-minus gluon tree amplitudes,' a finding previously thought impossible. This demonstrates a shift from LLMs regurgitating training data to performing novel scientific reasoning. Physicist Alex Lupsasca found that while GPT-5's general skills seemed stagnant, its frontier capabilities exploded, reproducing a complex paper in 11 minutes. This suggests expert 'priming' can unlock high-level reasoning in foundation models for compl
OpenAI's GPT-5.2 derived a new theoretical physics result concerning 'single-minus gluon tree amplitudes,' showing a gluon interaction can occur where physicists thought it couldn't. This highlights the 'jagged frontier' of AI capabilities: while general-use improvements in models like GPT-5 seemed minor, their capacity for high-level scientific reasoning surged. Physicist Alex Lupsasca, now at OpenAI, found GPT-5 could reproduce one of his best papers in 30 minutes. Using a 'priming' technique—solving a textbook problem first—the model then reproduced a result from a paper published after its training cutoff in just 11 minutes. This leap from calculation to novel derivation signals that foundation models are becoming tools for scientific discovery, not just information retrieval, requiring domain expertise to unlock their full potential.
Read the original → Latent Space
- #ai
- #llm
- #openai
- #gpt-5
- #research
Get five bites like this every day.
Tezvyn delivers a daily feed of 60-second tech bites with quizzes to lock in what you learn.