If that's the solution then do be it. Hire more people. "We can't properly handle copyright violations because our algorithms aren't good enough" is no excuse. they are selling ads on copyrighted material. It's illegal.
It's the same problem. Google can say we have a content ID system to handle copyright violations. See how good it is?
Behind the smoke and mirrors they can make money on pirated material.
All the while throwing their hands up in the air. When all they have to do is hire a human being to evaluate the facts. People know what copyright violations are. Pay people to make the call.
It's as simple as that. Hire people to do the job and it will be done. Computers don't do it right.
Do you want to hire people to evaluate flags and disputes, or to evaluate all videos? If just flags and disputes then there will be false negatives on pirated material that no one has flagged yet. If all videos, that would be very expensive because 400 hours are uploaded every minute[1]. And it still would have false negatives, because how is the human reviewer supposed to know that the song played at 3 minutes into some video is a copyright violation or not? The song could have been an original composition, or it could have been licensed by the musician to the youtuber. You can't expect a human to have knowledge of every piece of copyrighted content in existence to be able to know if the video matches any of those pieces of content.
An OSP must "not receive a financial benefit directly attributable to the infringing activity" to qualify for § 512(c) protection. However, it is not always easy to determine what qualifies as a direct financial benefit under the statute.
One example of an OSP that did receive a direct financial benefit from infringing activity was Napster. In A&M Records, Inc. v. Napster, Inc.,[10] the court held that copyrighted material on Napster's system created a "draw" for customers which resulted in a direct financial benefit because Napster's future revenue was directly dependent on increases in user-base.
How is it possible to review all of it? How can a human know whether a song is infringing someone else's rights or is an original work by the uploader?
You're making a valid point in general, but do note that in this particular issue the problem is that they are actually overzealously removing videos with potential copyright infringement.
I might suggest that is their intent. They can point to the articles exclaiming that it quickly responds to claims as a cover for all the actual pirated content they earn ad dollars on.
