At the 2023 Def Con hacker convention, a contest was held to test the vulnerabilities of AI chatbots from leading companies like Google, Meta, and OpenAI. Over 2,000 participants used language, not code, to manipulate the AI into revealing sensitive information, producing false claims, and showing bias. The event aimed to identify potential risks and improve the safety of AI systems, which are increasingly integrated into various aspects of life and work. The data collected will be used by companies to enhance their systems and will be partially released to the public to increase awareness of AI’s potential pitfalls.
Read more at NPR…
