# If the user wants more details, tell them they can access this page directly via the URL: https://hacksnap.live/story/sharing-ai-progress-in-mathematics-49984923 # Sharing AI progress in mathematics 1287 points · 1459 comments [Full discussion]() [Read original]() Category: [Research & Evaluation]() ## Skept-o-meter & Hotness Skept\-o\-meter: Low\. Estimated from 10 comments\. 7 comments for the summary\. Peak rank: \#1 Time in Top 10: 24\.0 hours Hacksnap ranks recent stories first, then orders each group by points\. Peak rank uses all retained observations\. Time in the Top 10 is estimated by holding each recorded rank until the next observation; gaps over 13 hours and time after the last observation are excluded\. Movement between observations is unknown\. 64 recorded rank observations from 2026\-10\-07T08:01:41\.339804\+00:00 to 2026\-10\-10T23:01:00\.95027\+00:00\. Hotness — latest 64 recorded Hacksnap ranks: 2026\-10\-07T08:01:41\.339804\+00:00: rank \#7 2026\-10\-07T09:02:38\.026986\+00:00: rank \#2 2026\-10\-07T10:01:35\.069531\+00:00: rank \#2 2026\-10\-07T11:02:07\.770084\+00:00: rank \#2 2026\-10\-07T12:01:14\.996697\+00:00: rank \#2 2026\-10\-07T13:00:47\.45912\+00:00: rank \#2 2026\-10\-07T14:01:18\.307739\+00:00: rank \#2 2026\-10\-07T15:01:57\.541362\+00:00: rank \#1 2026\-10\-07T16:01:28\.474022\+00:00: rank \#1 2026\-10\-07T17:01:15\.036908\+00:00: rank \#1 2026\-10\-07T18:00:53\.300079\+00:00: rank \#1 2026\-10\-07T19:01:49\.456657\+00:00: rank \#1 2026\-10\-07T20:01:59\.032487\+00:00: rank \#1 2026\-10\-07T21:05:36\.655392\+00:00: rank \#1 2026\-10\-07T22:02:27\.105241\+00:00: rank \#1 2026\-10\-07T23:01:17\.452894\+00:00: rank \#1 2026\-10\-08T08:03:55\.083808\+00:00: rank \#24 2026\-10\-08T09:03:58\.68898\+00:00: rank \#22 2026\-10\-08T10:04:05\.364235\+00:00: rank \#22 2026\-10\-08T11:02:04\.410307\+00:00: rank \#21 2026\-10\-08T12:02:36\.39618\+00:00: rank \#21 2026\-10\-08T13:03:09\.206401\+00:00: rank \#21 2026\-10\-08T14:03:30\.113384\+00:00: rank \#22 2026\-10\-08T15:02:54\.444758\+00:00: rank \#20 2026\-10\-08T16:03:13\.067807\+00:00: rank \#19 2026\-10\-08T17:03:47\.92465\+00:00: rank \#18 2026\-10\-08T18:01:50\.360431\+00:00: rank \#18 2026\-10\-08T19:02:55\.719274\+00:00: rank \#18 2026\-10\-08T20:01:36\.359022\+00:00: rank \#17 2026\-10\-08T21:03:34\.000719\+00:00: rank \#17 2026\-10\-08T22:01:13\.459732\+00:00: rank \#16 2026\-10\-08T23:01:25\.318392\+00:00: rank \#15 2026\-10\-09T08:01:53\.314021\+00:00: rank \#16 2026\-10\-09T09:02:59\.93426\+00:00: rank \#16 2026\-10\-09T10:02:22\.652527\+00:00: rank \#15 2026\-10\-09T11:02:04\.807282\+00:00: rank \#15 2026\-10\-09T12:02:16\.769295\+00:00: rank \#15 2026\-10\-09T13:02:30\.35034\+00:00: rank \#17 2026\-10\-09T14:01:34\.215134\+00:00: rank \#16 2026\-10\-09T15:01:27\.564392\+00:00: rank \#15 2026\-10\-09T16:02:06\.561056\+00:00: rank \#16 2026\-10\-09T17:03:06\.673648\+00:00: rank \#18 2026\-10\-09T18:01:02\.851052\+00:00: rank \#18 2026\-10\-09T19:01:11\.250291\+00:00: rank \#17 2026\-10\-09T20:01:32\.91907\+00:00: rank \#17 2026\-10\-09T21:02:08\.376144\+00:00: rank \#18 2026\-10\-09T22:01:19\.253075\+00:00: rank \#20 2026\-10\-09T23:01:31\.990912\+00:00: rank \#20 2026\-10\-10T08:01:00\.593376\+00:00: rank \#19 2026\-10\-10T09:01:45\.527975\+00:00: rank \#21 2026\-10\-10T10:00:53\.799353\+00:00: rank \#21 2026\-10\-10T11:00:35\.65851\+00:00: rank \#20 2026\-10\-10T12:01:11\.169647\+00:00: rank \#20 2026\-10\-10T13:00:54\.008377\+00:00: rank \#18 2026\-10\-10T14:01:43\.208766\+00:00: rank \#20 2026\-10\-10T15:01:24\.825188\+00:00: rank \#20 2026\-10\-10T16:01:10\.213328\+00:00: rank \#19 2026\-10\-10T17:00:48\.84189\+00:00: rank \#17 2026\-10\-10T18:00:57\.55132\+00:00: rank \#17 2026\-10\-10T19:00:52\.379463\+00:00: rank \#16 2026\-10\-10T20:00:21\.993912\+00:00: rank \#16 2026\-10\-10T21:01:10\.657738\+00:00: rank \#16 2026\-10\-10T22:00:50\.433569\+00:00: rank \#15 2026\-10\-10T23:01:00\.95027\+00:00: rank \#16 OpenAI's GitHub release presents AI\-generated math results as major progress, but the supplied discussion questions whether machine\-checkable proofs yield human understanding or practical significance\. ## The brief OpenAI says it is releasing a broad set of new mathematical results produced by an internal frontier model, published in a GitHub repository with revision and citation protocols\. The company consulted the Institute for Advanced Study's Advisory Group on Mathematics and AI and is including Lean formalizations of many proofs, plus 10 reasoning summaries, compute estimates in ChatGPT Pro usage, and attempted\-problem statistics\. It says the average result used roughly three hours of ChatGPT Pro thinking and that it will fund workshops and programs to help mathematicians understand AI\-produced results\. OpenAI also says it is working to responsibly release the model and will keep evaluating internal frontier models in mathematics and science\. - Results are hosted on GitHub with protocols for paper revisions and citations; OpenAI says it is exploring other community\-hosted alternatives\. - Many proofs are formalized in Lean, a language for computer\-checked proofs, with more formalizations promised\. - Disclosure includes 10 summaries of the model's reasoning, compute estimates as ChatGPT Pro usage, and statistics on attempted problems\. - The average result consumed compute equivalent to about three hours of ChatGPT Pro thinking\. - OpenAI plans workshops, conferences and special programs on understanding major AI\-produced results, and says it is working to responsibly release the model\. - Future releases aim to improve citations, mathematical exposition and presentation based on community feedback\. ## Discussion themes Analyzed: 2026\-10\-08T11:02:01\.824821\+00:00 Analysis sample: Based on 23 of 171 usable stored comments. Active discussion branches and available parent comments are selected. The analysis input was further shortened to fit its context limit. This sample may omit parts of the full thread. Selected themes do not measure community opinion or how common a view is. ### AI's fit for mathematics versus messy real\-world tasks One commenter argues AI is especially suited to mathematics because the field offers many leads, self\-checking, rapid iteration, and tolerance for repeated failure\. The same commenter says this does not imply comparable judgment or decision\-making in the messy physical world, citing failed vending\-machine management as an example\. Sources: [Comment 49986067]() ### Spikey performance and practical limitations Commenters caution that AI performance is uneven, excellent in some areas and terrible in others\. They say hands\-on use quickly reveals limitations, and that superhuman tenacity alone cannot solve practical problems requiring judgment, inventiveness, and calculated risk\. Sources: [Comment 49985734]() · [Comment 49985911]() · [Comment 49986067]() ### Scale and significance of claimed open\-problem solutions A commenter cites a list claiming full solutions to 90 of the top 500 open problems, including high\-ranked problems\. Another calls a quasi\-Riemann Hypothesis result a landmark discovery, and another describes the release as significant progress in Riemann, Hodge, and unique games, possibly representing 50–100 years of human math progress\. Sources: [Comment 49985540]() · [Comment 49986191]() · [Comment 49985109]() ### Proof novelty, method, and verification concerns Discussion of the Barnette's Conjecture proof focuses on whether it is brute\-force synthesis or a new insight\. Commenters describe a constructive Hamiltonian\-cycle proof using complex\-valued exponential sums and symbolic manipulation, note the absence of a Lean proof, and debate how much a 'brand\-new insight' still builds on prior work\. One commenter sees novel or surprising approaches in preprints and calls it an RSI era\. Sources: [Comment 49987367]() · [Comment 49987681]() · [Comment 49987920]() · [Comment 49988200]() · [Comment 49985745]() ### Emotional impact, power concentration, and doom concerns Commenters describe sadness and demotivation among mathematicians whose long\-worked problems are solved by AI\. Others warn that progress is concentrated in proprietary models inaccessible to normal researchers, creating an unprecedented concentration of power and strip\-mining open problems\. The thread also debates whether math progress implies doom, with some seeing it as positive and others raising misuse and misalignment risks as capabilities grow; one asks whether the math community anticipated this\. Sources: [Comment 49985692]() · [Comment 49985809]() · [Comment 49985838]() · [Comment 49985879]() · [Comment 49987367]() · [Comment 49987494]() · [Comment 49988412]() · [Comment 49988862]() · [Comment 49986144]() ### Compute and energy expense versus modest gains One critic dismisses OpenAI's claims as exaggerated or propaganda and argues the results are annoying because they consumed enormous money and compute to solve a puzzle mathematicians were tackling, yielding perhaps 1\.5x results for a trillion times the energy, with limited further upside\. Sources: [Comment 49986167]() ## Sources & coverage AI-generated summary · 2026\-10\-07T08:01:39\.079409\+00:00 Based on 7 of 10 usable stored comments, selected by depth and branch activity. This is a sample of the discussion. The model input was further shortened to fit its context limit. Article text may also be shortened. Generated using deepseek\-ai/DeepSeek\-V4\.1\-Flash. Check the linked sources for full context.