'Manipulative and disgraceful': OpenAI's critics seize on math benchmarking scandal

Fortune

Center

Factuality

Ownership

'Manipulative and disgraceful': OpenAI's critics seize on math benchmarking scandal

Reports suggest OpenAI is about to make a huge announcement about AI agents. But its last big announcement is already looking slightly questionable.

weirdreads.com

Factuality

Ownership

‘Manipulative and disgraceful’: OpenAI’s critics seize on math benchmarking scandal - WeirdReads

Reports suggest OpenAI is about to make a huge announcement about AI agents. But its last big announcement is already looking slightly questionable. Read Moresource The post ‘Manipulative and disgraceful’: OpenAI’s critics seize on math benchmarking scandal appeared first on WeirdReads.

Talouselämä

Factuality

Ownership

Mathematicians developed a test to really put new AI models to the test. Then it was revealed that the test was funded by an AI giant.

pivot-to-ai.com

Factuality

Ownership

OpenAI o3 beats FrontierMath — because OpenAI funded the test and had access to the questions

OpenAI released its o3 model in December, bragging about the model’s unparalleled ability to do math and science problems. The model’s success on the FrontierMath benchmark — solving 25.2% of…