It took OpenAI's agents 130 billion tokens to crack a 90-year-old math problem (www.businessinsider.com)

OpenAI has announced that its AI agents successfully solved the Navier-Stokes equations, a complex mathematical problem that has stumped mathematicians for nearly a century. This achievement required a staggering amount of computational resources, utilizing about 10,000 concurrent agents that communicated through 2.7 million messages over a span of 88 hours, followed by an additional 17 hours for verification. The effort is estimated to have cost between $10 million and $40 million, prompting OpenAI CEO Sam Altman to quip about an "AI bubble." The significance of this breakthrough is profound for the AI/ML community, as the Navier-Stokes equations are crucial for understanding fluid dynamics, with implications across various scientific fields and industries. However, the announcement has also sparked controversy, particularly regarding the use of research data from mathematicians Tristan Buckmaster and Levent Alpöge, who questioned OpenAI's reliance on their work. While OpenAI claims it did not directly use their data, the incident raises important discussions about data usage and attribution in AI training, highlighting both the potential of AI to tackle complex problems and the ethical considerations that accompany such advancements.