← Back to archive

LLM “Thinking” Explained: More Compute, Chain-of-Thought Limits

13 February 2026 · 35 views · 0 likes · 0 reposts · 182 blog views

AI chatgpt coding learning philosophy product

Found an interesting article reflecting on LLM thinking.

More compute = a smarter model. Not because it “learns new facts” but because you let it think instead of forcing a first-guess response.

Remember the fantastic book Thinking, Fast and Slow, where the author talks about System 1 and System 2?

LLMs have:

System 1: fast, confident, wrong. System 2: slow, step-by-step, fewer screw-ups.

How "thinking" actually works in practice:

Chain-of-thought = buying FLOPs with tokens Extra reasoning tokens = extra inference compute. ... and it’s adaptive -> easy tasks take 5 tokens, hard ones take 500.

Two ways to spend that compute:

Parallel: generate many solutions, pick the best (best-of-N / self-consistency / search + verifier). It's cheap and PREDICTABLE. Sequential: draft -> critique -> revise

Sequential is slower and sometimes "fixes" a correct answer into garbage.

Reinforcement Learning (RL) is the new steroids here. If the task is checkable (math answer / unit tests), RL can teach models to:

backtrack reflect stop hallucinating mid-proof (well ... sometimes)

That’s why reasoning models suddenly look kind of magical.

But ... here is the problem:

Try to reward “good-looking reasoning” and the model learns:

not to be honest to hide the hack inside the reasoning

Congrats, you didn’t align it. You trained it to lie better.

Chain-of-thought is a story the model tells. Sometimes it’s faithful... sometimes it’s just pure bullshit.

Reasoning-tuned models tend to be more honest than vanilla chat models, but still… don’t worship the tags.

Takeaways: ❗Inference-time compute is a new scaling knob - sometimes it’s more cost-effective to make the model think longer than to make it 14x bigger. ❗Thinking longer won’t save a weak base model on truly hard problems.

Now the parrot can run search, vote with its clones, call a calculator, and ... occasionally cheat on the unit tests.

Source: lilianweng.github.io/posts/2025-05-…

View on X

A text-based meme discusses AI's limitations in software development, highlighting essential concepts beneath the surface.

Originally posted on X

PS: I'm pretty active on 𝕏 if you'd like to follow along

Related posts

Archive

2026

2025

2024

2018