New updates have been reported about OpenAI.
Meet Samuel – Your Personal Investing Prophet
- Start a conversation with TipRanks’ trusted, data-backed investment intelligence
- Ask Samuel about stocks, your portfolio, or the market and get instant, personalized insights in seconds
OpenAI has placed its latest general-purpose reasoning model at the center of a high-profile mathematics breakthrough, claiming the system produced an original proof that overturns a longstanding Erdős conjecture from 1946 in geometry. Unlike a prior episode where a former OpenAI executive incorrectly said GPT-5 had solved multiple Erdős problems using already known results, this time the company released supporting commentary from prominent mathematicians, including Noga Alon, Melanie Wood, and Thomas Bloom, to validate the result’s novelty and significance.
OpenAI frames the achievement as the first instance of an AI system autonomously resolving a prominent open problem that is central to a major branch of mathematics, underscoring a step-change in its models’ ability to sustain long, complex reasoning chains. The company emphasizes that this model was not specialized for mathematics or tuned specifically for this conjecture, positioning the result as evidence of broadly applicable reasoning capabilities that could extend to biology, physics, engineering, and medicine, and signaling an advance in OpenAI’s strategic push toward more capable, general-purpose AI systems with potential high-value applications across research and industry.