Companies can be held responsible for their automated systems running amok. If this is the case, then either YT's ContentID system is overly sensitive, the third party uploaded something like a sample of white noise or silence that ContentID matched to the silence or jet engines in the video, or the third party uploaded the original author's video to ContentID.

If Google can be accused of defamation for automatically suggesting "sucks" as an additional search term after a company name, then they can be accused of defamation for automatically indicating that an original author's video contains another's content. If the third party deliberately uploaded ambiguous content, hoping to trigger spurious ContentID matches, that is fraud.

At the end of the original article, the author says the situation has been rectified, but this is not an isolated incident. The article links to another example of a third party uploading someone else's royalty-free audio tracks to ContentID, giving them ad revenue from any video that includes those audio tracks.

No third party should have this kind of unchecked power over others' creations, and YouTube's ContentID system has been upheld as an example of how all sites with user-generated content should operate.