https://samharris.org/episode/SE5DD11091B
Sam Harris speaks with Stuart Russell and Gary Marcus about recent developments in artificial intelligence and the long-term risks of producing artificial general intelligence (AGI). They discuss the limitations of Deep Learning, the surprising power of narrow AI, ChatGPT, a possible misinformation apocalypse, the problem of instantiating human values, the business model of the Internet, the meta-verse, digital provenance, using AI to control AI, the control problem, emergent goals, locking down core values, programming uncertainty about human values into AGI, the prospects of slowing or stopping AI progress, and other topics.
This is a great conversation about the challenges with generative AI as they exist 2023. There were two main points I took away from the conversation:
- The design of generative AI models like ChatGPT, while impressive, cannot reason about the world, which seriously limits the trust we can place in them.
- This design limitation makes it extremely unlikely that large language models will take us to the kind of robust general intelligence we need to solve important problems in the world.