They were also designed to addict adults, just saying.

Right, but adults are assumed to be somewhat more responsible for themselves. This is why we don't let kids (legally) smoke or drink, but we do let adults do so. We expect that adults can, in general, say no, and that children are less able to do so.

But it's not absolute. Some drugs are illegal for adults as well, for example. Why? Because they're too addicting.

So are Instagram and Youtube just nicotine, or are they heroin?

If you type in “Hi, how are you?” and go back and forth between English and LinkedIn Speak a few times, it winds up with “I’d like to sell you something.” Pretty accurate.

> Cope and cognitive dissonance

To be fair, AI probably wrote the blog post from a short prompt, which would explain the lack of detail.

This is 100% the case.

I wonder how much of the site was AI-generated. The images definitely are (kids with different numbers of fingers from each other in the same picture lol).

I really want the "Note from the Founder" to be fully AI generated too, image and everything. Our fully unauthentic web of the future has finally arrived!

It's not like before AI we got a lot of genuine founder stories either ;)

You could just have read the "about" page. Who cares if there's AI involved, this is a dad who made a thing for his own kids, and opened it up for everyone else. So what if they used AI if it does what they hoped it'd do and their kids like it?

Fine for them, but why would I use it instead of using my own AI for my own kids?

It's like reading other people's chat gpt conversations, not very interesting or useful.

> So what if they used AI

It matters because it's a very strong signal of quality.

I'd say it's a strong signal of (lack of) effort, which then correlates with quality

And, crucially, the quality _is_ low. At a bare minimum, the generated safety bulletpoints should be able to anticipate safety concerns with following the generated activities. The people publishing the posts should read them first and check for safety concerns.

The generated posts don't meet this bare minimum. For example, some posts have activities for toddlers involving string or rope, but do not mention the non-obvious strangulation risks. This website should not have been published in this state.

Because, if you AI generated the activities too, the site is useless. We can ChatGpt generate activities too and get the same result.

"You'll eat shit and like it" as a marketing strategy? I thought they fixed that in ChatGPT.

Quality, security, and code that doesn't fall apart later matter. I don't want AI slop children books to be a thing. I hear you on AI making it easier for people to build stuff, but calling valid concerns "gatekeeping" is a bit off.

With that said, I really like this site and the approach!

If those things actually mattered they would be rewarded in the "free marketplace of ideas". But they aren't and never have been. The most wealthy companies in the world aren't wealthy because of the quality of their code. That's why literally zero Very Large Organizations prioritize code quality. Marketing matters far more than quality and the budgets demonstrate this. You're just holding it wrong.

You're missing the point. Some slice of the market ignoring quality doesn't make it unimportant. Those companies get burned by tech debt and security holes all the time. Brushing off quality and security as pointless is shortsighted.

The study linked in OP is already a clear counterexample to your point, though. It's clear from all the slop that quality control is low on the priority list of so-called "builders" using AI. They do the first 20% to get a mockup and then decide it's done.

Your posts in this thread have said exactly what I was thinking

Same here about distrusting pieces that use AI art. It’s like… a signifier of something hard to describe — values, qualities — that makes me want to listen to someone’s point of view less.

It's a sign of low effort, maybe? If they're not putting much effort into the art, maybe they're not putting much into the writing, either. Maybe not even into the thinking.

This has to be parody, right?

I try to always work on something that’s contributing to something I care about in the world. I find that motivation is less of an issue if there’s intrinsic value to the work I’m doing.

For me, the "bloat" issue is less significant than the way that Electron apps don't have native macOS controls. They feel like a web app in a window, and attempts to emulate macOS-style controls just feel like facsimiles.