AI Safety Research directions: An Opinionated Ranking
Details
What do AI safety researchers work at? Ruben Castaing will briefly explain all of mech interp, AI control, model welfare, evals, compute verification, algorithmic bias, and more. Each of these is a field of study aiming to make AI go better, but they all have different theories as to what matters, and what we can and should investigate. Learn all about the different directions in AI safety research — and how Ruben ranks them against each other.
Afterwards we'll have an open discussion with drinks/food at a nearby bar.
### Pre-Reading and Resources
https://www.aisafetyguide.org/challenges
https://shallowreview.ai/similarity
Related topics
Events in Christchurch, NZ
Artificial Intelligence
Machine Learning
Risk Management
Risk Governance and Compliance
