A group of over 150 mathematicians has issued the 'Leiden Declaration on AI and Mathematics,' cautioning governments against the exaggerated claims regarding AI's capabilities in solving complex mathematical problems. The declaration, led by Ulrike Tillmann, vice president of the International Mathematical Union, emphasizes that while AI technologies like OpenAI's ChatGPT have made headlines for purportedly solving longstanding mathematical conjectures, their reliability is questionable. Experts warn that current AI models can produce convincing yet incorrect solutions, which poses a significant risk to ongoing mathematical research that relies on accurate prior results. The statement also highlights the troubling trend of mathematicians feeling pressured to endorse AI due to lucrative opportunities in the tech industry, amidst an era of underfunding in academia. Beyond mathematics, the declaration calls for regulatory oversight of AI, citing ethical concerns related to military use, misinformation, and environmental costs. This declaration serves as a critical reminder of the need for careful scrutiny of AI's role in research and its broader societal implications.