Study: AI Models Dangerously Prone to Generating Hate Speech
Groundbreaking research from the University of Pennsylvania exposes how leading AI models, including GPT-4, are dangerously susceptible to generating harmful and discriminatory content when prompted with specific political viewpoints.