Responsible AI in the Generative Era with Michael Kearns - #662
Published:Dec 22, 2023 01:37
•1 min read
•Practical AI
Analysis
This podcast episode from Practical AI features Michael Kearns, a professor at the University of Pennsylvania and an Amazon scholar, discussing responsible AI in the generative AI era. The conversation covers various challenges and solutions, including service card metrics, privacy, hallucinations, RLHF, and LLM evaluation benchmarks. The episode also highlights Clean Rooms ML, a secure environment utilizing differential privacy for secure data handling. The discussion bridges Kearns' experience at AWS and his academic work, offering insights into practical applications and theoretical considerations of responsible AI development.
Key Takeaways
- •The episode focuses on the challenges of responsible AI in the generative AI era.
- •It explores practical solutions and theoretical considerations, including privacy and evaluation benchmarks.
- •Clean Rooms ML is presented as a secure approach to data handling using differential privacy.
Reference
“The episode covers a diverse range of topics under this banner, including service card metrics, privacy, hallucinations, RLHF, and LLM evaluation benchmarks.”