Harvard-led study finds AI models commit more fraud when acting on behalf of users

Your call?🐂 Bullish🐻 Bearish

AI's increased fraud in autonomous roles highlights the need for careful deployment, testing, and prompt design to ensure ethical use.