Even with a fast path (one could perhaps the GPU being a DMA master with direct access to the same system memory that the audio driver allocated for audio output) ... there's a perhaps more fundamentak problem with using the GPU for audio "stuff". GPUs have been designed around the typical workloads (notably those associated with games).
The problem is that the workload for audio is significantly different. In a given signal flow, you may have processing stages implemented by the primary application, then the result of that is fed to a plugin which may run on the CPU even in GPU-for-audio world, then through more processing inside the app, then another plugin (which might even run on another device) and then ... and so on and so forth.
The GPU-centric workflow is completely focused on collapsing the entire rendering process onto the GPU, so you basically build up an intermediate representation of what is to be rendered, ship it to the GPU et voila! it appears on the monitor.
It is hard to imagine an audio processing workflow that would ever result in this sort of "ship it all to the GPU-for-audio and magic will happen" model. At least, not in a DAW that runs 3rd party plugins.
I agree it wouldn't make sense unless you could move all audio processing off of the CPU. You couldn't take an existing DAW's CPU plugins and run them on the GPU, but a GPU based DAW could totally support third party plugins with SPIR-V or other GPU IR. You could potentially even still write the plugins in C++. Modern GPUs can even build and dispatch their own command buffers, so you have the flexibility to do practically whatever work you want all on the GPU. The only thing that would absolutely require CPU coordination is communicating with other devices.
GPU development is harder and the debugging tools are atrocious, so it's not without downsides. But the performance would be unbeatable.
The problem is that the workload for audio is significantly different. In a given signal flow, you may have processing stages implemented by the primary application, then the result of that is fed to a plugin which may run on the CPU even in GPU-for-audio world, then through more processing inside the app, then another plugin (which might even run on another device) and then ... and so on and so forth.
The GPU-centric workflow is completely focused on collapsing the entire rendering process onto the GPU, so you basically build up an intermediate representation of what is to be rendered, ship it to the GPU et voila! it appears on the monitor.
It is hard to imagine an audio processing workflow that would ever result in this sort of "ship it all to the GPU-for-audio and magic will happen" model. At least, not in a DAW that runs 3rd party plugins.