Will the largest machine learning training run (in FLOP) as of the end of 2030 be in the United States?
Mini
11
Ṁ171
2031
74%
chance

As of 2023, machine learning (ML) models have been consuming increasing amounts of computational power for training, with FLOPs (floating-point operations per second) used as a measure of these resources. The location of the largest training run can be influenced by several factors, including access to hardware, energy costs, and regulatory environment.

As of December 31, 2030, will the largest machine learning training run (in terms of FLOP) have occurred in the United States?

Resolution Criteria:

This question will resolve positively if, by December 31, 2030, there is credible evidence that the largest machine learning training run, measured in FLOP, occurred in the United States. Evidence must come from a credible and verifiable source such as a recognized media outlet, a government report, an academic paper, or a company announcement.

This question will resolve negatively if by the given date, the largest known machine learning training run occurred outside the United States or if no credible evidence has been provided indicating that the largest run occurred in the United States.

If a report is released but is later retracted or debunked by a credible source, the question will also resolve negatively.

In the event of conflicting reports, the resolution will be based on the consensus of credible sources. If no consensus is achieved, the question will resolve as ambiguous.

Get Ṁ1,000 play money
Sort by:

What if it's trained in a distributed way?