Frontier AI labs drove recent trader focus when OpenAI and Anthropic announced in late October 2026 that internal large language models had produced claimed refutations of the Exponential Time Hypothesis, including explicit subexponential-time algorithms for k-SAT problems. Both firms released preprints and Lean-formalized proofs shortly after their models generated the arguments in mid-October, prompting an open review led by ETH co-originators Russell Impagliazzo and Ramamohan Paturi alongside Ryan Williams. Independent Lean rebuilds have succeeded and toy-model checks align, yet experts emphasize that full verification of the statements against standard ETH formulations remains incomplete, with no peer-reviewed confirmation as of early November. Ongoing community scrutiny, including planned virtual meetings and expositions, represents the key near-term catalyst that could solidify or challenge the market-implied odds.