From my perspective, this is nothing more than hype marketing, because "Our product is too good to release it yet" is something I have seen many times before.

For instance, back before the release of the PlayStation 2 (!!!), it was claimed that the high-performance Emotion Engine in the PS2 could be used for weapons guidance systems. - Source

It is also interesting—and enlightening—that the same security exploits could be found using smaller (and cheaper) open-source models.

Nobody: Enshitification?
Anthrophic: Hold my Spyware.

Here is how platforms die: first, they are good to their users; then they abuse their users to make things better for their business customers; finally, they abuse those business customers to claw back all the value for themselves. Then, they die.

I had a lot of love for GitHub Copilot in the past, but you cannot stop enshittification.

This app has a lot of potential. It's like Numi, it's like my scrappad on the right bottom corner of my Mac, it's math, it's a to-do list, it's cool. And only $5.

I've noticed the same trend Paul Kinlan describes. We're at a point where React isn't just the most popular JavaScript framework anymore — it's become its own platform.

New frameworks are "dead on arrival." React has become the platform because LLMs generate React by default. It's a self-reinforcing loop: more React in training data → more React in output. For LLMs, a prompt like "build me a website" is basically synonymous with "build me a React app."

That said, I've found that Vue.js and Nuxt work just fine too — though you occasionally need to point the LLM back to the documentation. The model doesn't default to them, but it can be guided.

React's top position will likely stay for a long time. And honestly? That makes me a bit sad. It means fast-paced frontend development is probably going to slow down significantly. We shouldn't expect many major innovations in the near future.