State of AI in 2026: LLMs, Coding, Scaling Laws, China, Agents, GPUs, AGI | Lex Fridman Podcast #490
17:08extended thinking. Extended thinking and
inference-time scaling is just a way to make the models marginally smarter.
And I will always48:06- I guess the big question here is, we talked
quite a bit about the architecture behind the pre-training. Are the scaling laws
holding strong across pre-training,48:16post-training, inference, context
size, data, synthetic data? - I'd like to start with the technical
definition of a scaling law—














