The company that created ChatGPT has just shared a massive update on its scientific research efforts. The team at OpenAI recently announced that its internal models have successfully produced three major breakthroughs across the fields of mathematics, computer science, and theoretical physics. Alongside these massive claims, the organization is rolling out a completely different strategy for how it works with outside experts to verify and publish its findings.
Models tackle longstanding math and computer science problems
The organization shared that its artificial intelligence models recently proved the local smoothing conjecture. This specific math proof apparently resolves the well-known Kakeya conjecture at the same time. To back up the math, the team published the proof in a machine-readable format called Lean so anyone can check the work right now.
In the computer science world, the system also claims to have proven the Exponential Time Hypothesis false. This hypothesis is a big deal in theoretical computing. Instead of just putting the paper online, the team handed the proof over to an independent group of researchers from top universities to review the work first.
Don’t miss the best of The Mac Observer
Set us as a preferred source and our Apple reporting ranks higher in your Google Search results and Discover feed — one tap, no account changes.
External experts now guide the publication of new findings
This brings up the new academic approach. For the computer science and physics results, the company is bringing in external collaborators known as shepherds. These independent scientists volunteer to review the model outputs, verify the claims, and figure out the best way to release the materials to the public.
- The experts do not sign any nondisclosure agreements.
- The reviewers can publish their own conclusions freely.
- The company will not take academic credit for the discoveries.
For its third big discovery regarding quantum chromodynamics in physics, a group of these shepherds is currently deciding how and when to release the research materials.
New funding helps researchers study these massive breakthroughs
To make sure independent scientists actually have the time and resources to look into these AI breakthroughs, financial support is also part of the plan. A new funding initiative managed by the Simons Foundation will pay for special programs, conferences, and workshops.
This money ensures researchers can properly examine, explain, and build upon the new discoveries. It marks a clear shift from private tech labs simply posting papers on the internet to actually helping the academic community verify the unproven work.
Discussion