![\"Kevin](\"https:\/\/www.newsbeep.com\/ca\/wp-content\/uploads\/2025\/10\/openai-researcher-deletes-tweet.jpg\")
<\/p>\n\t<\/p>\n
\t\t\tIllustration by Tag Hartman-Simkins \/ Futurism. Source: Getty \/ X (Screenshots)<\/p>\n
An OpenAI exec bragged that ChatGPT had discovered a mathematical breakthrough \u2014 and was immediately left with egg on his face when it turned out to be bogus, The Decoder reports<\/a>.<\/p>\n We begin where any good controversy begins: with a now-deleted tweet<\/a>. This one was by Kevin Weil, a vice president at OpenAI who last week proclaimed that the company\u2019s newest large language model, GPT-5, had \u201cfound solutions to 10 (!) previously unsolved Erd\u0151s problems and made progress on 11 others.\u201d\u00a0<\/p>\n The Erd\u0151s problems refer to a number of tricky mathematical conjectures made by the Hungarian mathematician of the same name. If the claims are true, this would be a notable achievement for an AI, and an awesome demonstration of the ChatGPT\u2019s purported \u201cPhD level\u201d intelligence.<\/p>\n But the mathematician who literally runs the website erdosproblems.com<\/a>, a researcher at the University of Manchester named Thomas Bloom, tweeted<\/a> that this was \u201ca dramatic misrepresentation,\u201d because the AI had only found existing work that already solved the problems.\u00a0<\/p>\n In fact, the problems weren\u2019t even \u201cunsolved\u201d in the first place. They were listed as \u201copen\u201d on the Erd\u0151s problems website, which Bloom stressed only meant that he personally was unaware of a paper with a solution, not that no solution exists.<\/p>\n \u201cNo need to describe it as something it\u2019s not!\u201d Bloom wrote.<\/p>\n OpenAI\u2019s competitors blasted Weil for the blunder, per TechCrunch<\/a>. Google DeepMind CEO Demis Hassabis called<\/a> it \u201cembarrassing.\u201d Hyperbolic Labs cofounder Yuchen Jin said<\/a> that the episode calls\u00a0 for \u201cbetter peer review for these \u2018AI discovers science\/math\u2019 claims.\u201d And the notoriously blunt Yann LeCun, Meta\u2019s chief AI scientist, summed it up with a brutal joke: \u201cHoisted by their own GPTards.\u201d<\/p>\n It\u2019s another example of the careless boosterism of OpenAI and the tech industry at large. To hype the launch of GPT-5, OpenAI repeated the claim that the AI had achieved \u201cPhD-level intelligence<\/a>,\u201d even though it\u2019s still often incapable of giving a straight answer<\/a> to basic questions. CEO Sam Altman warns-slash-hypes that the tech is getting good enough to replace entire categories of jobs<\/a>, while also declaring that the company is on the verge of building an artificial general intelligence, or AGI, that roundly outperforms humans in every domain.<\/p>\n Science and mathematics in particular is one way that AI companies are trying to hijack some credibility. Maybe AI isn\u2019t perfect yet, but who can argue with the endeavor if widely available chatbots are pushing the boundaries of empirical pursuits? Elon Musk, for instance, claims that his \u201cmaximum truth-seeking<\/a>\u201d chatbot Grok will discover<\/a> \u201cnew technologies\u201d and\u00a0\u201cnew physics,\u201d if not the \u201ctrue nature of the universe.\u201d<\/p>\n While some scientists have found uses for generative AI, it\u2019s largely as a research tool, dredging up the scientific literature it devoured in its training that may otherwise get buried in the results of a search engine. But anytime someone from the industry claims that chatbots are coming up with breakthroughs on their own should be taken with a grain of salt. In this case, all ChatGPT seems to have done is copy someone else\u2019s homework.<\/p>\n