OpenAI claimed to solve the Navier-Stokes problem, a $1-million Clay Mathematics Institute prize, but mathematicians argue it solved a variant with an artificial external force rather than the fundamental problem the field cares about. Three mathematicians proved OpenAI's approach cannot extend to the full problem, and experts contend the company exploited a loophole in the prize's original formulation.