Well, of course it does. “The debate over AI governance is stuck in a false choice between over-regulation and no regulation. There is a middle way,” noted the tech giant in a blog post about its latest report.
“A pragmatic, evidence-based approach that recognizes the unique challenges and opportunities of both frontier AI and widely-deployed AI applications.”
Google, like other tech firms and music-industry bodies, has been energetically lobbying policymakers in the US over What They Should Do About AI for some time now. Its new report sets out some of the key things that it’s lobbying for.
The section on ‘Creativity, copyright, and the AI value exchange’ is the one Music Ally readers will jump to, and in some cases quickly bristle at.
“Using publicly available web data for training models is a transformative, non-expressive use — like an art student taking inspiration from walking through a gallery — that should remain protected under fair use in the U.S. and text-and-data-mining exceptions abroad,” claims the report.
It’s an argument that can be pushed back on, because there is payment and permission involved in art galleries, even the ones that are free to enter from the visitor’s perspective.
If there is revenue from sales of the art, it’s shared with the artists, and where works have been bequeathed by an artist or their estate, they did it willingly – the gallery didn’t just wander into their house and take them.
But let’s continue. Google’s report then calls for AI companies to “recognize rights holders’ desires to find beneficial pathways for creative professionals and knowledge workers”.
This comes with a reference to its own deals “in which we are paying for access to and delivery of diverse types of specialized, non-public content”, which are cited as the kind of partnerships “most likely to be sustainable for developers and meaningful for the ecosystem”.
The report also makes it clear that Google’s other policy ask is that “in addressing copyright concerns, the focus should again be on outputs — in this case, whether a specific image or piece of text actually copies an existing work, regardless of how it was created”.
This is interesting, so we’ll quote it in full:
“Technical safeguards can help prevent models from generating outputs that reproduce works they were trained on. However, to protect the open space needed for new forms of creative expression, it is important that filters do not try to automate subjective decisions like whether something is ‘too similar’ to a prior work,” claims Google.
“Here, the appropriate mechanism is through established notice and-action frameworks that use standard reporting and takedown mechanisms to remove infringing content.”
More bristling in the music industry. Based on Deezer’s latest stats, there are now 75,000 fully AI-generated tracks being uploaded every single day to streaming services – up from 10,000 a day at the start of 2025.
Rightsholders will be dubious over whether standard reporting and takedown mechanisms can really deal with the sheer scale of AI music – let along other forms of AI-generated content – that’s being made and put out. And yes, a Content ID-style system may be the answer, but Google clearly has skin in the game with that suggestion.
Lobbying is lobbying: Google’s report merely sets out the arguments that it’s making in private to policymakers around the world. But for the music industry, the publication is another reminder to sharpen their own arguments in response.
