AI benchmarking organization criticized for waiting to disclose funding from OpenAI

A corporation creating math benchmarks for AI didn’t disclose that it had obtained funding from OpenAI till comparatively not too long ago, drawing allegations of impropriety from some within the AI group.

Epoch AI, a nonprofit primarily funded by Open Philanthropy, a analysis and grantmaking basis, revealed on December 20 that OpenAI had supported the creation of FrontierMath. FrontierMath, a take a look at with expert-level issues designed to measure an AI’s mathematical expertise, was one of many benchmarks OpenAI used to demo its upcoming flagship AI, o3.

In a post on the discussion board LessWrong, a contractor for Epoch AI going by the username “Meemi” says that many contributors to the FrontierMath benchmark weren’t knowledgeable of OpenAI’s involvement till it was made public.

“The communication about this has been non-transparent,” Meemi wrote. “In my opinion Epoch AI ought to have disclosed OpenAI funding, and contractors ought to have clear details about the potential of their work getting used for capabilities, when selecting whether or not to work on a benchmark.”

On social media, some users raised considerations that the secrecy might erode FrontierMath’s status as an goal benchmark. Along with backing FrontierMath, OpenAI had entry to most of the issues and options within the benchmark — a reality Epoch AI didn’t disclose previous to December 20, when o3 was introduced.

In a reply to Meemi’s submit, Tamay Besiroglu, affiliate director of Epoch AI and one of many group’s co-founders, asserted that the integrity of FrontierMath hadn’t been compromised, however admitted that Epoch AI “made a mistake” in not being extra clear.

“We have been restricted from disclosing the partnership till across the time o3 launched, and in hindsight we should always have negotiated tougher for the flexibility to be clear to the benchmark contributors as quickly as attainable,” Besiroglu wrote. “Our mathematicians deserved to know who may need entry to their work. Although we have been contractually restricted in what lets say, we should always have made transparency with our contributors a non-negotiable a part of our settlement with OpenAI.”

Besiroglu added that whereas OpenAI has entry to FrontierMath, it has a “verbal settlement” with Epoch AI to not use FrontierMath’s downside set to coach its AI. (Coaching an AI on FrontierMath could be akin to teaching to the test.) Epoch AI additionally has a “separate holdout set” that serves as an extra safeguard for unbiased verification of FrontierMath benchmark outcomes, Besiroglu stated.

“OpenAI has … been totally supportive of our determination to keep up a separate, unseen holdout set,” Besiroglu wrote.

Nonetheless, muddying the waters, Epoch AI lead mathematician Ellot Glazer noted in a post on Reddit that Epoch AI hasn’t be capable to independently confirm OpenAI’s FrontierMath o3 outcomes.

“My private opinion is that [OpenAI’s] rating is legit (i.e., they didn’t prepare on the dataset), and that they haven’t any incentive to lie about inside benchmarking performances,” Glazer stated. “Nonetheless, we are able to’t vouch for them till our unbiased analysis is full.”

The saga is yet another example of the problem of creating empirical benchmarks to judge AI — and securing the required assets for benchmark growth with out creating the notion of conflicts of curiosity.