"We made a mistake in not being more transparent": OpenAI secretly accessed benchmark data, raising questions about the AI model's supposedly "high scores" …
A new report suggests OpenAI secretly funded and accessed the FrontierMath benchmarking data, raising concerns about whether ...
AI benchmarking organization criticized for waiting to disclose funding from OpenAI
An organization developing math benchmarks for AI didn't disclose that it had received funding from OpenAI until relatively ...
‘Manipulative and disgraceful’: OpenAI’s critics seize on math benchmarking scandal
In this edition…Watch OpenAI’s hands; Trump scraps Biden’s AI order; Whistleblower targets Amazon and Covariant; Titans vs.
OpenAI Secretly Funded Benchmarking Dataset Linked To o3 Model
OpenAI secretly funded and had access to a benchmarking dataset, raising questions about high scores achieved by its new o3 ...
Digital information world2d
AI Benchmark Group Under Fire for Delayed Disclosure of OpenAI Funding
Contributors criticize FrontierMath’s undisclosed OpenAI funding, highlighting secrecy, transparency issues, and ethical ...
AI Models Are Getting Smarter. New Tests Are Racing to Catch Up
A new set of much more challenging evals has emerged in response, created by companies, nonprofits, and governments. Yet even on the most advanced evals, AI systems are making astonishing progress. In ...
OpenAI Just Pulled a Theranos with o3: Transparency Questions Emerge
OpenAI just pulled a Theranos with o3 by claiming record-breaking performance on the FrontierMath benchmark while having ...
GIGAZINE1mon
Mathematicians talk about the shock of OpenAI's o3 model scoring 25.2% on the ultra-difficult math dataset 'FrontierMath'
Kevin Buzzard, a mathematician and professor of pure mathematics at Imperial College London, posted a blog post explaining how OpenAI's o3 model scored 25.2% on the FrontierMath problem dataset.