A federal judge has sharply questioned the Trump administration's justification for labeling artificial intelligence company Anthropic a national security risk, stating the government has failed to provide sufficient evidence to support its ban on federal use of the company's technology.