Why your speculative and very biased duplicate post should be of any concern for RME?

Ok, i may watch too many youtube videos and therefore tend to overstate things a bit ;-)
But anyway, lets see what the future brings to the table.

Arno

BTW, DL algorithms don't need some half-baked anonymous home studio project to train on when they can use commercially available music releases for the same task with greater success.

And no, the AI assistants built into some OS do not work in the way the OP believes they do.

Dear friends,

There are several logical issues in the claims presented.

First, the idea that unreleased music is “useless” for AI training is technically incorrect. Machine learning models do not learn from popularity; they learn from structure: harmony, rhythm, sound design, and production techniques. For a model, diversity and volume of data matter far more than Spotify chart positions. Being unreleased does not make data worthless.

Second, the claim that “Apple would never allow the abuse of customer data” is an absolute statement and therefore logically unsound. No commercial entity can guarantee “never.” Apple strongly markets privacy, but it has also revised practices in response to past criticism. A closed ecosystem does not automatically equal perfect transparency.

Third, concerns about Windows are not entirely unfounded. Windows 11 includes telemetry by design, and Microsoft Copilot operates at least partially through cloud-based processing. Additionally, the widely discussed Recall feature—which periodically captures screenshots of the user’s screen for “memory” purposes—faced significant security criticism and was delayed and redesigned. This demonstrates that privacy and security concerns are tied to concrete architectural decisions, not paranoia.

Fourth, Android is not a binary issue. There are de-Googled Android variants and alternative mobile operating systems such as Ubuntu Touch. Not every Android user is automatically accepting full Google data integration.

In summary, the comments relies heavily on generalizations and absolute statements without technical nuance. A critical perspective on AI features, telemetry, and operating system security is not conspiracy thinking—it is normal risk assessment in a rapidly evolving technological landscape.

If you want to jump in at the deep end, you can watch.
www.youtube.com/@robbraxmantech

Thank you for clarifying your point.

However, the reason we do not want our music to be freely used for AI training is simple: copyright law protects our work. Our music is our intellectual property, and it cannot be used without permission.

It is not about fear of influence or about hiding something. It is about respecting the rights of the creators. Just because something is good does not mean it should automatically be available for any kind of use — especially commercial AI training — without consent.

If someone wants to use our music for learning or training purposes, the proper way is to ask and agree on terms.

Respect for copyright is respect for the artist.

Beyond copyright, this is also about privacy and boundaries in general.

When boundaries are crossed gradually, people are slowly conditioned to accept more and more. At first, it seems harmless. Then it becomes normal. Eventually, things that would have clearly felt wrong at the beginning are no longer questioned.

It is a bit like the “boiling frog” metaphor: if you put a frog into hot water, it jumps out immediately. But if you place it in cold water and slowly raise the temperature, it does not notice the danger until it is too late.

The same can happen with privacy and rights. Small compromises may not seem significant on their own, but over time they can fundamentally change what we consider acceptable.

That is why it is important to draw clear lines early — not out of paranoia, but out of principle.

I understand your frustration, but your post oversimplifies the issue. Saying that AI learning from art is “just like human inspiration” ignores the scale and nature of AI training: machines can ingest entire catalogs of copyrighted works in ways a person cannot, and use them for commercial purposes without consent.

Your point about public domain works like Handel is valid, but that doesn’t justify conflating modern copyrighted recordings or performances with centuries-old compositions. Respecting copyright is not about stopping creativity — it’s about ensuring creators retain control over how their work is used.

Finally, labeling any concern as “FUD” dismisses legitimate discussions about ethics, legality, and the practical impact of AI on artists. Criticism doesn’t equal paranoia.