As a parent, I think people overestimate the average quality of pre-AI children’s content.

Also interesting to look beyond children’s work at the incredible amount of “book mill” content that has dominated publishing for hundreds of years.

We rightly celebrate the good ones, but most content before AI was not good. So not surprising that the AI trained on that corpus is of similar quality.

That being said, I think this is just an opportunity to improve AI content, which is a human/computer interface design challenge. This technology is here to stay. Our focus should be on detection and improving the LLMs.

I’ve had luck formalizing this into some post-LLM rules to clean crappy default AI content before I work with it: https://slopwash.com

I think that generative AI has many more downsides than upsides. It is and will continue to be a net negative for society, unless we have the collective discipline to manage the dangers.

We're long past the point of these being the dark nefarious actions of a hidden cabal and you're doing a psa alerting us to a conspiracy.

Consumers are paying for this and every other "terrible" thing being done "to them."

The incorrect place to put blame is the company trying to get away with it. It's the consumers that are literally paying for this stuff.

This is probably bad but as another comment noted most people don’t understand how bad it is right now. I’ll give you an example: you might’ve heard the word “Cocomelon,” because it’s one of the most popular children’s television programs in the last 25 years. It’s entirely on YouTube so for a long time, we didn’t actually know who made Cocomelon. This was a show that was a household word among people who had toddlers, and we had no fucking clue who was behind it until the Wall Street Journal found out.

It’s just a total mess out there.