IP Policy Institute Study Shows Significantly Higher Rates of Section 101 Invalidations for AI Patents

University at Buffalo Professor Amy Semet's study analyzing 14,000 AI patents litigated in U.S. district courts from 2000-2025 finds AI patents face significantly higher Section 101 subject matter eligibility invalidation rates than non-AI patents, particularly post-Alice v. CLS Bank (2014). Conversely, AI patents show lower obviousness invalidations under Section 103. The research reveals AI patents face a "double hurdle" - harder to keep valid and harder to enforce when valid, creating systematic disadvantages for AI inventions in patent litigation.