Thinking – what is something it? What does it mean for people to think? Are we about to be overtaken? artificial do you think Many people think so, but not Teppo Felin, as I understand him. In this episode, EconTalk host Russ Roberts welcomes back Felin to discuss these questions, based on a working paper Felin wrote with Matthias Holweg. Felin explains that this is a human ability don’t pay attention Existing data and evidence is not only our Achilles heel, but also one of our greatest strengths. So what are we actually do if we think???
After listening to the podcast, we’d love to hear your thoughts. Share your answers to the commands below in the Comments, or use them to start your conversation offline. Either way, we’re here for you.
1- What is wrong with thinking of the brain as a computer? This metaphor has been around since the 50s. It’s true, says Felin, that neural networks make connections, but the brain isn’t quite a computer yet—there’s something else going on. How should we understand the brain instead, according to Felin?
2- Large Language Models can take in a lot of content, Felin says it would take people hundreds of years to pre-train their brains with the same number of words. He describes LLMs as using a stochastic process that is very good at “forecasting the future.” If LLMs contain so much, why can’t they think like people? What is the importance of hindsight in the human thought process?
3- Felin says that using AI to make rational decisions is a fool’s game, as is the idea of being able to remove biased AI. Why is bias a feature, not a bug, of human cognition? What is the relationship of belief and theory to thinking, and how does this differentiate human thinking from AI?
4- Felin uses the good example of the Wright brothers, emphasizing that no Venture Capitalist could give them money. Why, according to Felin, AI would have affected development of the Wright brothers? Why, on the contrary, were they able to succeed?
5- Roberts quotes Yann LeCun; “Prediction is the backbone of intelligence.” How much prediction corresponds to intelligence, according to Felin? What does Felin mean when he says, “…that’s something computers can’t do. They take existing data as given, while we humans discover and create, experimentally, new data.” How much difference does it make between digital and biological intelligence, and how would you explain it?
Source link