“A study by researcher Fabrizio De La Cua found that skilled individuals using capable AI systems tend to "fall asleep at the wheel," overly trusting the AI's output without applying sufficient critical oversight.”