“A significant problem with AI model benchmarks is data contamination, where models are inadvertently trained on the evaluation data, leading to artificially high scores that do not reflect true generalization capabilities.”