Monday, November 17th at 7pm, I'll be at Harvard Bookstore with particle physicist and author Daniel Whiteson, co-author of "Do Aliens Speak Physics?"

Rapid progress in artificial intelligence (AI) capabilities has drawn fresh attention to the prospect of consciousness in AI. There is an urgent need …

If we want to stay at the forefront of knowledge production, we must fit technology to our needs.

“We survived, but it wiped out the library,” Internet Archive’s founder says.

The article argues against the notion that large reasoning models (LRMs) cannot think, countering Apple's claim that they merely perform pattern-matching. It refutes the idea that LRMs fail because they struggle with complex algorithmic tasks, drawing a parallel to human limitations. The author posits that LRMs almost certainly can think, defining thinking in terms of problem-solving and linking it to brain functions like problem representation, mental simulation, pattern matching, monitoring, and insight. Chain-of-thought reasoning in LRMs is compared to inner speech in humans, with LRMs demonstrating backtracking capabilities. The author states that next-token prediction, the basis of LRMs, is a powerful form of knowledge representation, aligning with the expressive power of natural language. A next-token predictor requires world knowledge and the ability to represent intermediate steps, which proves that it is thinking. Even though a perfect auto-complete would seem omniscient, LRMs meet the criteria to represent the thinking process. Open-source LRM benchmarks show successful problem-solving, demonstrating reasoning abilities, even outperforming untrained humans in certain instances. The author concludes that LRMs possess thinking capabilities by meeting the requirements to do so across the board.

Research from Anthropic on the ability of large language models to introspect

The official home of the Python Programming Language

YouTube video by Jia-Bin Huang