GPT-5.2

In my experience, the best models are already nearly as good as you can be for a large fraction of what I personally use them for, which is basically as a more efficient search engine.

The thing that would now make the biggest difference isn't "more intelligence", whatever that might mean, but better grounding.

It's still a big issue that the models will make up plausible sounding but wrong or misleading explanations for things, and verifying their claims ends up taking time. And if it's a topic you don't care about enough, you might just end up misinformed.

I think Google/Gemini realize this, since their "verify" feature is designed to address exactly this. Unfortunately it hasn't worked very well for me so far.

But to me it's very clear that the product that gets this right will be the one I use.

> It's still a big issue that the models will make up plausible sounding but wrong or misleading explanations for things, and verifying their claims ends up taking time. And if it's a topic you don't care about enough, you might just end up misinformed.

Exactly! One important thing LLMs have made me realise deeply is "No information" is better than false information. The way LLMs pull out completely incorrect explanations baffles me - I suppose that's expected since in the end it's generating tokens based on its training and it's reasonable it might hallucinate some stuff, but knowing this doesn't ease any of my frustration.

IMO if LLMs need to focus on anything right now, they should focus on better grounding. Maybe even something like a probability/confidence score, might end up experience so much better for so many users like me.

I ask for confidence scores in my custom instructions / prompts, and LLMs do surprisingly well at estimating their own knowledge most of the time.

I’m with the people pushing back on the “confidence scores” framing, but I think the deeper issue is that we’re still stuck in the wrong mental model.

It’s tempting to think of a language model as a shallow search engine that happens to output text, but that metaphor doesn’t actually match what’s happening under the hood. A model doesn’t “know” facts or measure uncertainty in a Bayesian sense. All it really does is traverse a high‑dimensional statistical manifold of language usage, trying to produce the most plausible continuation.

That’s why a confidence number that looks sensible can still be as made up as the underlying output, because both are just sequences of tokens tied to trained patterns, not anchored truth values. If you want truth, you want something that couples probability distributions to real world evidence sources and flags when it doesn’t have enough grounding to answer, ideally with explicit uncertainty, not hand‑waviness.

People talk about hallucination like it’s a bug that can be patched at the surface level. I think it’s actually a feature of the architecture we’re using: generating plausible continuations by design. You have to change the shape of the model or augment it with tooling that directly references verified knowledge sources before you get reliability that matters.

You have a subtle slight of hand.

You use the word “plausible” instead of “correct.”

That’s deliberate. “Correct” implies anchoring to a truth function the model doesn’t have. “Plausible” is what it’s actually optimising for, and the disconnect between the two is where most of the surprises (and pitfalls) show up.

As someone else put it well: what an LLM does is confabulate stories. Some of them just happen to be true.