The ML-based similar sound search in Live 12 has been a huge help. It replaced searching for the right percussion sound with a button on the drum rack that moves through similar sounds at an instrument or rack level. I can also use it to search for similar Foley to add variety in the textures that tie the track together.

It's a black box that takes human produced artwork without consent and spits out superficial mediocre content a dime a dozen. It also takes away developer time and focus from other aspects of the software. I don't think you need to understand the algorithms underneath to have a problem with that.

I think it's important to contextualize the situation. The Adobe TOS were updated, and people where understandably concerned over the rights Adobe now has over private customer data. The biggest issue was this part: "Licenses to Your Content. Solely for the purposes of operating or improving the Services and Software, you grant us a non-exclusive, worldwide, royalty-free sublicensable, license, to use, reproduce, publicly display, distribute, modify, create derivative works based on, publicly perform, and translate the Content."

People looked at this and immediately assumed this was added to allow Adobe to train models with people's private work.

Adobe has now updated their TOS, but this was a breach of trust.

Either way, in the end, this potential AI threat is just another reason to not store stuff in the "cloud".

> You say that but in my experience the same kinds of people who will say that are also the kinds of people who demonstrably don't actually even know what AI is.

Funny. In my experience AFK it is the people who use AI that have zero idea what it is and think answers can be blindly trusted. The ones who don’t like it can enumerate the drawbacks clearly.