OpenAI launches FrontierScience, a benchmark to measure models' expert-level scientific reasoning with 700+ questions, finding GPT-5.2 is its strongest model (OpenAI) Main Link | Techmeme Permalink
1 repost10 likes
OpenAI launches FrontierScience, a benchmark to measure models' expert-level scientific reasoning with 700+ questions, finding GPT-5.2 is its strongest model (OpenAI) Main Link | Techmeme Permalink
1 repost10 likes
stop calling it "reasoning" because it can't. bsky.app/profile/symb...
chatgpt and llm based toys CANNOT REASON, researchers conclude, confirming what everyone has already observed, except liars and frauds & every tech billionaire. #aifraud #openai #agi #sellsellsell machinelearning.apple.com/research/gsm...