It's just a super simple skill that, when invoked, makes the model spend considerable time asking design and architecture questions and fleshing out any plan with you. A planning session without it might be Claude asking you 2 questions, and with it 22.

You're right. There is technically a category of work that relies on neither our ability to do physical labor nor excessive thinking. It just relies on being a human.

The conclusion is thus obvious: AI is going to push us all into careers as photo models, OF-creators, and social media influencers! /s

These 0.6B-4B models are, frankly, just amusing curiosities. But commonly regarded as too error prone for any non-demo work.

The reason why people are buying Apple Silicon today is because the unified memory allows them to run larger models that are cost prohibitive to run otherwise (usually requiring Nvidia server GPUs). It would be much more interesting to see benchmarks for things like Qwen3.5-122B-A10B, GLM-5, or any dense model is the 20b+ range. Thanks.

It's easy to forget that us here on HN are several standard deviations from the norm.

I was more so limiting myself to the simpler heuristic where people only pay roughly what they personally think something is worth, and not significantly more/less regardless of the options. But of course, as you've pointed out, in real life the options available really do matter, and someone might decline a 200:1200 trade if there are even more lopsided options available. It does complicate the though experiment somewhat if you try to take this into account.

Of course this doesn't take into account people who just pay to play around and learn, non professional use cases, or a few other things, but it's a rough ballpark estimate.

Assuming the above, current AI models would only increase the productivity for most workplaces by a relatively small amount, around 10-200 € per employee per month perhaps. Almost indistinguishable compared to salaries and other business expenses.

On one hand I agree with you. Any automatic filter implemented can later be expanded to cover more and more things, such as messages from political adversaries for example. It's a slippery slope as we all know.

On the other hand I don't think it applies in this context very much. If we're talking about content published by a corporation or such (say a newspaper for example) they already filter all their gathered news themselves and have no obligation to publish things they don't feel like.

Similarly if we're talking about user uploaded content on social media I don't think they have any obligation to publish everything and anything that their users decide to upload either, and it's not the expectations of the users that anything can be hosted there for them. Users already know that youtube/facebook/tiktok/what-have-you have seemingly arbitrary rules regarding what content they're willing to host and not.

Now if for example DNS providers or ISPs decide to implement these sort of filters on the web at large that's a different matter I think. In which case I agree with you.

How brutal will the enshittification phase of these products be?

Will the 10x cost or whatever be something that future employers will have to pay, or will it be a more visible impact for all of us? Assuming no AGI scenario here and the investments will have to be paid back with further subscription services like today.

I really hope Open Source (Open Weights) keep up with the development, and that a continuation of Moore's Law (the bastardized performance per € version) makes local models increasingly accessible.

It's very difficult to train for that. Of course you can include a Question+Answer pair in your training data for which the answer is "I don't know" but in that case where you have a ready question you might as well include the real answer anyways, or else you're just training your LLM to be less knowledgeable than the alternative. But then, if you never have the pattern of "I don't know" in the training data it also won't show up in results, so what should you do?

If you could predict the blind spots ahead of time you'd plug them up, either with knowledge or with "idk". But nobody can predict the blind spots perfectly, so instead they become the main hallucinations.

Of course, there are other issues instead.

I agree with you, this is rather odd. And sort of missing the problem.

All apps are about attention. The percentage of the time spent using the app when it shows you your good content (Whatever it is that you're interested in) determines how addictive it is. And the percentage of time it's showing you bad content (Ads, 'screen time breaks', manual scroll time, more ads, loading screens, sponsor ads, filler content (youtube for instance is full of this), etc) counteracts the addictive properties because nobody likes it.

What's the end goal here? Right now TikTok is winning the attention economy race against the other apps because it's more focused on the user's preferred content. Is that what we want to reduce? To show more uninteresting other stuff on the screen? Like blank 'wait 5 minute' statics? Or just more ads?

I get that we don't want a generation of socially inept phone addicts, but this won't solve anything I fear. People will still want the good content, forcing the most customer friendly (it feels wrong to say that about TikTok) app to become more enshittified is a bewildering solution.

But this really just stinks of Regulatory Capture to me. Their main argument is that the consumers like to use the app too much?

Why? Because it's smarter and not as enshittified as the competitors?

I'm sure if youtube, facebook, reddit, etc reduced the number of ads, and started showing more relevant content that people actually cared about, they too would start being "more addictive". Do we really want to punish that?

What's the end goal here?

More plausibly: You registered the domain. You created the webpage. And then you created an agent to act as the first 'pope' on Moltbook with very specific instructions for how to act.

> AI assistance produces significant productivity gains across professional domains, particularly for novice workers. Yet how this assistance affects the development of skills required to effectively supervise AI remains unclear. Novice workers who rely heavily on AI to complete unfamiliar tasks may compromise their own skill acquisition in the process. We conduct randomized experiments to study how developers gained mastery of a new asynchronous programming library with and without the assistance of AI. We find that AI use impairs conceptual understanding, code reading, and debugging abilities, without delivering significant efficiency gains on average. Participants who fully delegated coding tasks showed some productivity improvements, but at the cost of learning the library. We identify six distinct AI interaction patterns, three of which involve cognitive engagement and preserve learning outcomes even when participants receive AI assistance. Our findings suggest that AI-enhanced productivity is not a shortcut to competence and AI assistance should be carefully adopted into workflows to preserve skill formation -- particularly in safety-critical domains.

We currently live in the very thin sliver of time where the internet is already full of LLM writing, but where it's not quite invisible yet. It's just a matter of time before those Dead Internet Theory guys score another point and these comments are indistinguishable from novel human thought.

Took me a few years to realize that this wasn't a universal feeling, and that many others found the programming tasks more fulfilling than any challenging engineering. I suppose this is merely another manifestation of the same phenomena.