The US accuses several Chinese artificial intelligence companies of maliciously copying technology from American AI firms. US law enforcement and intelligence officials say the firms used a technique called “distillation” to transfer knowledge from US-made AI models into their own products.
According to reporting, distillation works by training a model to replicate the outputs of another model, potentially allowing companies to benefit from intellectual property without direct code copying. US officials describe the activity as unlawful or improper and frame it as an intellectual property concern that could further complicate US–China relations.
Some outlets connect the accusation to broader diplomacy, noting it comes ahead of a planned meeting between leaders of the world’s two largest economies later this month. Others focus more narrowly on the technical mechanism—how training methods may be used to reproduce capabilities derived from US systems—rather than on the timing alone.
