[jcl]

I think the issue here is not that the DMCA has been abused, but that it has been circumvented entirely. Under the DMCA, the copyright holder would have to identify relevant works, decide if they think their copyright has been violated, then issue a takedown request (which can later be challenged by the work's author, at which point the work is reinstated and there is a court case).

Under YouTube's scheme, the copyright holder merely hands over a set of audio "signatures" and all works matching them are muted -- even if they may turn out to be fair use. The copyright holder never identifies infringing works directly, and there is presumably no chance to challenge the decision. That's not how the DMCA is supposed to work.

While YouTube's scheme is probably catching mostly copyright violations, it's kind of scary that they have equated "algorithmic signature matches" with "copyright violations". The reason "fair use" is such a gray area is because it's really hard to define what is free speech and what is a copyright violation. Handing the decision to an algorithm isn't a good solution.