[flagged]

> This is obviously untrue so either the author is knowingly lying or is plain incompetent. We had LLMs not being able to do simple high school mathematics a year back, now it is solving open problems in mathematics. Fields medalists in physics and mathematics are using it on a daily basis.

I'm confused, how do you think this disproves the claim in the article?

What do you think that quoted portion was talking about?

▲

simianwords 7 hours ago | parent [-]

The two claims are not related, my bad for making it seem so. I about that a serious person would still needs proof that AI is becoming more efficient. If you need proof, here's one:

> A year ago, we verified a preview of an unreleased version of @OpenAI o3 (High) that scored 88% on ARC-AGI-1 at est. $4.5k/task

> Today, we’ve verified a new GPT-5.2 Pro (X-High) SOTA score of 90.5% at $11.64/task

> This represents a ~390X efficiency improvement in one year

https://x.com/arcprize/status/1999182732845547795

This is not even including newest improvements. GPT 5.6 beats GPT 5.5 by 1 OOM.

https://x.com/GregKamradt/status/2075274981794300113

▲

unknownfuture 6 hours ago | parent [-]

That's still not the kind of efficiency the article is talking about.

They're referring to raw training and inference scaling and it's relationship (or lack thereof) to traditional economies of scale that we've seen with past technologies where they get cheaper as adoption increases, not more expensive.

Its true that the models requiring fewer turns and tokens due to increasing sophistication improves cost efficiency for users but that doesn't address the fundamental computational scaling problems of LLM architectures.

▲

simianwords 5 hours ago | parent [-]

This doesn't mean anything. Here are the facts

1. models themselves are getting cheaper - around 5000x in the past 1.5 years

2. ironically, if it were fully cheap, the same crowd would say that this would make AI bubble pop because where would they get moat?

3. training is also getting cheaper, the whole lifecycle to train and do inference on models from 1 year ago is on the whole 50x cheaper or so

I also think the author's point is a bit nebulous

> Chatbot companies are aware that their products are inefficient. Some have found techniques for improving performance, but they have not yielded significant gains

Performance has increased. To you specifically: what metric would falsify the claim that "they have not yielded gains" and "their products are inefficient"?

The same metric should apply to

- industries like steel

- pharma

- internet/cloud computing

- automobile

- consumer electronics

All of them have had high impact. So any criticism on inefficiency should be unique to LLM's and shouldn't apply to all of them. Please answer.

▲

unknownfuture 4 hours ago | parent [-]

All of those industries benefited and continue to benefit from economies of scale in the manufacturing and delivery of their products. You're not making the case you think you are, and I have a suspicion the article's point is just sailing right past you.

▲

simianwords 3 hours ago | parent [-]

notice how you didn't give me a metric that proves efficiency/inefficiency?

	▲	unknownfuture 2 hours ago \| parent [-]
		And now I know you either don't understand what an economy of scale is or you're not discussing in good faith. Carry on!