New research funded by the Pentagon suggests that artificial intelligence can scan and analyze blocks of text to discern whether the humans who wrote it have done something wrong or not.
Pentagon-Funded Study Uses AI to Detect ‘Violations of Social Norms’ in Text