OpenAI has rejected allegations that it used unpublished work from New York University mathematician Tristan Buckmaster in connection with its Navier-Stokes proof. In its latest response to the dispute, the company said an internal review found that prompts Buckmaster submitted to Codex over the past two months could not have influenced the internal model involved, including through training. The issue emerged after Buckmaster and Levent Alpöge, who had been working on the same line of research, entered large amounts of draft material into Codex. After their progress reached OpenAI, the company produced a Navier-Stokes proof within days using concentrated computing power, prompting Buckmaster to question whether his unpublished research had been absorbed by the model. OpenAI said that did not happen. The company also drew a firm cutoff at July 3, stating that no user input after that date could have affected the system. Some community members have speculated that the date may correspond to a training data cutoff or model checkpoint, but OpenAI did not provide an explanation. Buckmaster and Alpöge’s key breakthrough came in mid-August, well after the July 3 boundary cited by OpenAI.
OpenAI has issued a fresh response to the dispute over credit for a Navier-Stokes proof, explicitly denying that it used unpublished research from New York University mathematician Tristan Buckmaster.
After an internal review, the company said prompts Buckmaster submitted to Codex over the past two months could not have influenced the internal model in any way, including through training.
Buckmaster and Levent Alpöge had been working on the same research direction and had placed large volumes of draft material into Codex. After their progress reached OpenAI, the company used concentrated computing power and produced a Navier-Stokes proof within days. That led Buckmaster to question whether his unpublished work had already been learned by the model.
OpenAI has now directly rejected that claim. The company also set July 3 as a hard boundary, saying no user input after that date could have affected the system.
Some in the community have speculated that the date may mark a training data cutoff or a model checkpoint, but OpenAI did not explain it.
Buckmaster and Alpöge made their key breakthrough in mid-August, well after the July 3 boundary cited by OpenAI.
This article was originally published by Bit.Fan. For more cryptocurrency news and market insights, visit www.bit.fan. Disclaimer:
The market information, project data, and third-party content displayed on this platform are for industry information sharing only and do not constitute any form of investment advice or return commitment.
Cryptocurrency trading carries high risks. Users should fully assess their risk tolerance and make independent decisions. All profits, losses, and legal responsibilities are borne by the users themselves.
