Leonard Tang

Leonard Tang and Brian Huang from Haize Labs share their insights on AI model vulnerabilities and automated jailbreaking techniques. They discuss the crucial role of the o1 Red Team in testing OpenAI's latest reasoning models, emphasizing the balance between AI's advanced capabilities and potential risks. The conversation delves into automated red teaming strategies, the challenges of evaluating AI safety, and the ongoing battle between model functionality and security measures. Tune in for a deep dive into the future of AI technology and its implications!

Leonard Tang, Co-founder at Haize Labs, joins the conversation to discuss their innovative open source evaluation library, Verdict, aimed at improving AI judgment reliability. They dive into the fascinating capabilities of Grok 3, comparing its performance with competitors and addressing censorship challenges. Tang also highlights the impact of nepotism bias in AI models and how Verdict seeks to enhance evaluation efficiency. The podcast explores the exploratory advances in robotics, including robots learning to hand objects to one another, showcasing the exciting future of AI.