For our next event we have a guest speaker, Arush Tagade, speaking about evaluating the safety and capabilities of contemporary AI models!
This talk will present motivation behind model evaluations and jailbreaks along with its relevance to AI Safety. This will be followed by an overview of AI Safety careers. There will also be a Q&A at the end.
Arush is an interpretability and model evaluations researcher currently contracted with Leap Labs. He received his Masters from the University of Sydney in 2022 and worked for an AI hardware startup before transitioning to independent research in January 2023. He was a TA at ARENA 2.0 in May 2023, and has been a TA for Intro to ML Safety and AGI Safety Fundamentals in 2022.
