←back to thread

Jagged AGI: o3, Gemini 2.5, and everything after

(www.oneusefulthing.org)
265 points ctoth | 9 comments | 20 Apr 25 14:55 UTC | HN request time: 0.426s | source | bottom
Show context
fsmv ◴[20 Apr 25 17:31 UTC] No.43745151[source]▶
It's not AGI because it still doesn't understand anything. It can only tell you things that can be found on the internet. These "jagged" results expose the truth that these models have near 0 intelligence.

It is not a simple matter of patching the rough edges. We are fundamentally not using an architecture that is capable of intelligence.

Personally the first time I tried deep research on a real topic it was disastrously incorrect on a key point.

replies(4): >>43745177 #>>43745178 #>>43745251 #>>43745758 #
simonw ◴[20 Apr 25 17:34 UTC] No.43745178[source]▶
Is one of your personal requirements for AGI "never makes a mistake?"
replies(1): >>43745286 #
1. Arainach ◴[20 Apr 25 17:53 UTC] No.43745286[source]▶
I think determinism is an important element. You can ask the same LLM the same question repeatedly and get different answers - and not just different ways of stating the same answer, very different answers.

If you ask an intelligent being the same question they may occasionally change the precise words they use but their answer will be the same over and over.

replies(4): >>43745344 #>>43745362 #>>43745395 #>>43745545 #
2. hdjjhhvvhga ◴[20 Apr 25 18:01 UTC] No.43745344[source]▶
If determinism is a hard requirement, then LLM-based AI can't fulfill it by definition.
3. samuel ◴[20 Apr 25 18:06 UTC] No.43745362[source]▶
That's not an inherent property of the system. You can choose the most likely token(topk=1) and it will be deterministic (at least in theory, in some hardware setups it might be trickier)
4. beering ◴[20 Apr 25 18:09 UTC] No.43745395[source]▶
A human will give different answers to the same question, so I’m not sure why it’s fair to set a higher bar for an LLM. Or rather, I’m not sure how you would design this test in a way where humans would pass and the best LLM would fail.
5. simonw ◴[20 Apr 25 18:30 UTC] No.43745545[source]▶
That's because "intelligent beings" have memory. If you ask an LLM the same question within the same chat session you'll get a consistent answer about it.
replies(1): >>43745726 #
6. Arainach ◴[20 Apr 25 18:55 UTC] No.43745726[source]▶
I disagree. If you were to take a snapshot of someone's knowledge and memory such that you could restore to it over and over, that person would give the same answer to the question. The same is not true for an LLM.

Heck, I can't even get LLMs to be consistent about *their own capabilities*.

Bias disclaimer: I work at Google, but not on Gemini. If I ask Gemini to produce an SVG file, it will sometimes do so and sometimes say "sorry, I can't, I can only produce raster images". I cannot deterministically produce either behavior - it truly seems to vary randomly.

replies(3): >>43745943 #>>43746313 #>>43754715 #
7. IanCal ◴[20 Apr 25 19:31 UTC] No.43745943{3}[source]▶
You could run an llm deterministically too.

We're often explicitly adding in randomness to the results so it feels weird to then accuse them of not being intelligent after we deliberately force them off the path.

8. danielbln ◴[20 Apr 25 20:28 UTC] No.43746313{3}[source]▶
You'd need to restore more than memory/knowledge. You'd need to restore the full human, and in the exact same condition (inside and out).

Ask me some question before bed and again after waking up, I'll probably answer it at night but in the morning tell you to sod off until I had coffee.

9. int_19h ◴[21 Apr 25 18:06 UTC] No.43754715{3}[source]▶
Of course it varies randomly, that's literally what temperature is for in generation.