Abstract

The seminar will be on OpenAI’s new AI safety model, https://blog.openai.com/debate/ .

How can we augment humans so that they can effectively supervise advanced AI systems? One way is to take advantage of the AI itself to help with the supervision, asking the AI (or a separate AI) to point out flaws in any proposed action. To achieve this, we reframe the learning problem as a game played between two agents, where the agents have an argument with each other and the human judges the exchange. Even if the agents have a more advanced understanding of the problem than the human, the human may be able to judge which agent has the better argument (similar to expert witnesses arguing to convince a jury).

Bring a laptop if you can, so we can play the cat/dog game! https://debate-game.openai.com/