AI has witnessed remarkable strides in the realm of mathematics. Claude has successfully accomplished the formal proof of Fermat's Last Theorem. Additionally, whispers abound that it has unraveled the mysteries of millennium problems like the Navier-Stokes equations. However, these assertions remain unverified through direct confirmation. Nevertheless, AI has already carved out significant milestones across a spectrum of mathematical conundrums, such as elevating the proportion of zeros of the Riemann zeta function that align with the conjecture and resolving several mathematical puzzles that have long eluded solution. Shing-Tung Yau has put forth a reward of RMB 100,000 for a formal validation of Grigori Perelman's proof of the Poincaré conjecture. Terence Tao expresses concern that AI's black-box solution techniques might circumvent the gradual accumulation of problem-solving tools, potentially sapping mathematics of its vital growth impetus. Despite these apprehensions, AI developers persist in championing related contests, and market projections hint that AI could unlock millennium prize problems by the year 2026.