The second part of the chain is that you also tailor the operations the die area support to those operations necessary in a typical neural network, further optimizing the chip.
The end result is very power-efficient execution of neural networks, which allows you to beat a GPU or CPUs power curve, improving thermals of the core, and in the case of mobile devices, optimizes battery usage.
Those efficiency wins are almost certainly worth it even if third-party developers don't use it much.
Basically it's used by anything involving machine learning:
- Speech recognition
- Face recognition
- Visual lookup (image recognition)
- Live Text (OCR)
It allows all these functions to be performed efficiently on-device rather than shipping data off to the cloud.