我問ChatGPT: Did AI ever prove a mathematical theorem?這是它的回答:

Yes — AI has already been used to prove new mathematical theorems, though usually in collaboration with human mathematicians. A few examples:

 

 

1. DeepMind and Knot Theory (2019–2021)

 

 

Researchers at DeepMind worked with mathematicians to apply machine learning to knot theory and representation theory. The AI system spotted patterns in large datasets that led to new conjectures. Mathematicians then proved these conjectures rigorously, producing new theorems in pure mathematics.

 

 

2. Formal Proof Assistants

 

 

Systems like Lean, Isabelle, and Coq have been used to formally verify long and difficult proofs — for example, parts of the Feit–Thompson theorem and the Kepler conjecture. These don’t usually discover the theorem, but they ensure that proofs are correct beyond doubt. Increasingly, AI is being integrated into these systems to help automate proof steps.

 

 

3. Automated Theorem Provers (ATP)

 

 

Programs like E, Vampire, and Prover9 can prove many theorems automatically from given axioms. They’re especially strong in algebra, logic, and combinatorics. These are not “AI” in the modern deep-learning sense, but they are automated reasoning engines that sometimes find proofs too complex for humans to stumble upon easily.

 

 

4. Recent Advances (2023–2025)

 

 

  • Lean + GPT-style AI: Researchers have been training large language models to generate formal proofs inside Lean. These AIs can prove a growing portion of theorems from math libraries without human guidance.
  • Discovery: In some cases, AI has gone beyond checking proofs — it has generated new lemmas or proof strategies that human mathematicians hadn’t considered, nudging math toward machine-discovery.
  • High-profile result: In 2023, an AI-assisted effort produced a new result in geometry of polynomials, where the machine’s pattern recognition directly guided the conjecture and proof.

 

 

 

 

So the short answer:

AI has already helped prove existing theorems rigorously and, in a few cases, contributed to the discovery of entirely new theorems. The field is moving quickly, with hybrid human–AI collaboration producing the most exciting results.

 

Do you want me to give you concrete names of the new theorems AI helped discover, like the ones from DeepMind’s work on knot theory?

 

所有跟帖: 

ChatGPT更進一步敘述: -大觀園的賈探春- 給 大觀園的賈探春 發送悄悄話 大觀園的賈探春 的博客首頁 (35859 bytes) () 09/29/2025 postreply 18:02:07

所以不僅Coding正被AI化,數學證明也正被AI化。 -大觀園的賈探春- 給 大觀園的賈探春 發送悄悄話 大觀園的賈探春 的博客首頁 (0 bytes) () 09/29/2025 postreply 18:05:03

那哲學能自動化嗎? -角尺和規- 給 角尺和規 發送悄悄話 (0 bytes) () 09/29/2025 postreply 18:06:05

數理邏輯作為哲學的一個分支沒有理由不能AI化。 -大觀園的賈探春- 給 大觀園的賈探春 發送悄悄話 大觀園的賈探春 的博客首頁 (0 bytes) () 09/29/2025 postreply 18:07:22

AI現在除了識別圖像,語音,基本就是個search engine. 肯本沒有獨立思考能力。 -角尺和規- 給 角尺和規 發送悄悄話 (0 bytes) () 09/29/2025 postreply 18:11:23

你太低估AI了。 -大觀園的賈探春- 給 大觀園的賈探春 發送悄悄話 大觀園的賈探春 的博客首頁 (0 bytes) () 09/29/2025 postreply 18:12:21

就像你,用AI來證明AI. -角尺和規- 給 角尺和規 發送悄悄話 (0 bytes) () 09/29/2025 postreply 18:13:12

數學如此,物理還會遠嗎? -大觀園的賈探春- 給 大觀園的賈探春 發送悄悄話 大觀園的賈探春 的博客首頁 (0 bytes) () 09/29/2025 postreply 18:06:16

至於化學,去年的諾貝爾化學獎就是頒發給了研究AI的人。 -大觀園的賈探春- 給 大觀園的賈探春 發送悄悄話 大觀園的賈探春 的博客首頁 (0 bytes) () 09/29/2025 postreply 18:08:35

你來給大家科普一下 AI 到底證明了什麽重要定理, 重要在哪裏,in your own words -STEMkid- 給 STEMkid 發送悄悄話 (0 bytes) () 09/29/2025 postreply 18:14:00

上麵ChatGPT不是已經說了嗎? -大觀園的賈探春- 給 大觀園的賈探春 發送悄悄話 大觀園的賈探春 的博客首頁 (0 bytes) () 09/29/2025 postreply 18:26:53

看上去是做的是很簡單的事情,重要在哪呢?你看懂了嗎? -STEMkid- 給 STEMkid 發送悄悄話 (0 bytes) () 09/29/2025 postreply 18:30:00

起步嘛。 -大觀園的賈探春- 給 大觀園的賈探春 發送悄悄話 大觀園的賈探春 的博客首頁 (0 bytes) () 09/29/2025 postreply 18:39:51

再問問: “Can AI be used to fact check Trump?" -skyport- 給 skyport 發送悄悄話 skyport 的博客首頁 (0 bytes) () 09/29/2025 postreply 18:20:29

重點是協助, 協助數學家去證明定理。 -家有高中小娃- 給 家有高中小娃 發送悄悄話 (0 bytes) () 09/29/2025 postreply 19:41:12

請您先登陸,再發跟帖!