Recent OpenAI announcements of AI-generated partial proofs for Yang-Mills existence and a weaker quasi-Riemann hypothesis have positioned those outcomes as market leaders, reflecting trader views on accelerating large language model capabilities in formal mathematics. However, the 42% implied probability on no full solution by December 2027 captures widespread skepticism that verification, peer review, and complete resolutions will occur within the tight timeline, given the problems' depth and historical precedent for delays. Partial results on related conjectures like Hodge or Birch and Swinnerton-Dyer have drawn less attention, while P versus NP remains sidelined due to its computational complexity barriers. Upcoming expert shepherding of the claims could shift sentiment if full confirmations emerge.