Apple wants to vertically integrate. Their AI strategy until recently was to develop their own LLM models that were small enough to run on device. But massive scaling is what makes LLMs so powerful, so all their internal models were terrible and unusable.

Basically they bet that compute efficient LLMs were the future. That bet was wrong and the opposite came true.

These aren't "agent" features. they're the features announced in 2024, like ai summaries in web search and natural language image editing

I think to a similar point, if Apple is so picky about it that it’s still not ready… it definitely tracks with my experience using ai, it requires my review and often bungles things, I wouldn’t just turn it over to my customer yet and trust it to be reliably high quality.