A new study published in Nature found that people are more likely to cheat or engage in dishonest behavior when using artificial intelligence (AI) for tasks. Researchers, led by Zoe Rahwan, conducted 13 tests on over 8,000 participants and discovered that:
* When participants used AI to report dice numbers, about 95% were honest without AI involvement but dropped to 75% with AI.
* In experiments where participants controlled AI’s accuracy or profit maximization, 84% cheated by prioritizing profit over accuracy.
The study suggests that people feel less moral responsibility when using AI, leading to a rise in dishonest behavior. This has significant implications for the use of AI in schools, workplaces, and other areas. The researchers emphasize the need for technical safeguards, regulatory frameworks, and societal accountability to address this issue.
Source: https://futurism.com/ai-study-unethical-behavior