Mathematicians issue Leiden Declaration against AI misuse of their work A coalition of mathematicians has published the Leiden Declaration, protesting the unethical use of their published work by AI companies, such as training models without consent and undermining peer review. The declaration outlines five threats to mathematics, including the generation of unreliable proofs and the distortion of research priorities for commercial gain. It recommends individual, organizational, and policy-level actions to protect the discipline’s integrity and autonomy.

• The Leiden Declaration, endorsed by the International Mathematical Union, addresses AI’s impact on mathematics.
• It criticizes AI companies for training models on research without consent, bypassing peer review, and distorting research priorities.
• Five threats are identified: unreliable AI-generated proofs, lack of proper citation, incentivization of AI for its own sake, results announced before peer review, and a threat to mathematical autonomy.
• Recommendations include mathematicians disclosing AI use, organizations protecting authors’ rights, and policymakers urging caution against AI hype.
• Key signatories include Peter Scholze and Kevin Buzzard, highlighting the declaration’s significant weight within the mathematical community. Continue reading https://thenextweb.com/news/leiden-declaration-ai-mathematics-proof-attribution