Episode 6 – The Illusions of thinking
Controversy between AI giants! In this episode Arijit and Christian discuss Apple’s paper that suggested that Large Reasoning Models collapse at a certain level of complexity. This finding set the AI community alight and instigated an interesting rebuttal from Antropic that highlighted some amateur errors made by the Apple team. Even though errors may have been made there are important lessons to be learned for teams implementing LRM’s.
https://ml-site.cdn-apple.com/papers/the-illusion-of-thinking.pdf
https://arxiv.org/pdf/2506.09250v1
00:00 Introduction
01:09 Summary of the Apple paper
02:31 Understanding the Towers of Hanoi
04:44 Navigating complexity in problem solving
06:18 The role of reasoning in AI models
08:58 Performance discrepancies in AI models
09:46 Anthropic’s rebuttal and critique
12:44 Philosophical considerations of AI thinking
26:16 Compression Techniques: Lossy vs Lossless
15:52 Conculsions and future directions
18:52 Close out
