Posted on twenty eighth August 2024 by Harry
Arvind Narayanan is a professor of Pc Science at Princeton and the director of the Middle for Info Expertise Coverage. He’s a co-author of the guide AI Snake Oil and an enormous proponent of the AI scaling myths across the significance of simply including extra compute. He’s additionally the lead writer of a textbook on the pc science of cryptocurrencies which has been utilized in over 150 programs world wide, and an accompanying Coursera course that has had over 700,000 learners.
In Right now’s Episode with Arvind Narayanan We Focus on:
1. Compute, Information, Algorithms: What’s the Bottleneck:
- Why does Arvind disagree with the generally held notion that extra compute will end in an equal and steady stage of mannequin efficiency enchancment?
- Will we proceed to see gamers transfer into the compute layer in the necessity to internalise the margin? What does that imply for Nvidia?
- Why does Arvind not imagine that knowledge is the bottleneck? How does Arvind analyse the way forward for artificial knowledge? The place is it helpful? The place is it not?
2. The Way forward for Fashions:
- Does Arvind agree that that is the quickest commoditization of a know-how he has seen?
- How does Arvind analyse the way forward for the mannequin panorama? Will we see a world of few very massive fashions or a world of many unbundled and verticalised fashions?
- The place does Arvind imagine essentially the most worth will accrue within the mannequin layer?
- Is it attainable for smaller corporations or college analysis establishments to even play within the mannequin house given the extreme money wanted to fund mannequin improvement?
3. Training, Healthcare and Misinformation: When AI Goes Incorrect:
- What are the one greatest risks that AI poses to society right this moment?
- To what extent does Arvind imagine misinformation by way of generative AI goes to be an enormous drawback in democracies and misinformation?
- How does Arvind analyse AI impacting the way forward for schooling? What does he imagine everybody will get flawed about AI and schooling?
- Does Arvind agree that AI will be capable of put a health care provider in everybody’s pocket? The place does he imagine this principle is weak and falls down?
