We're pleased to report that Sony Music backtracked on its accusation of copyright infringement against the Hudson Valley Bluegrass Association, and HVBA's educational video remains freely available to the public. But the music label’s response leads us to think that Sony's misuse of copyright and of YouTube’s automated enforcement system will continue.

We wrote last week about how YouTube’s system, Content ID, incorrectly flagged HVBA’s own video as infringing. The video, an hour-long lecture on the history of bluegrass music, triggered the Content ID filters because it contained three clips of bluegrass recordings copyrighted by Sony, each around 30 seconds and surrounded by a discussion of the music and its historical relevance. That’s an obvious fair use under copyright law, one that any human reviewer with minimal training would recognize.

A fair use doesn’t require permission from the copyright holder, or a fee. It’s the sort of use that’s free to all. But when HVBA’s webmaster wrote to Sony Music and asked them to withdraw the Content ID match, the company responded by asking for a $500 “administrative fee” and detailed information about HVBA’s use of the song clips. Fortunately, HVBA’s webmaster knew her rights, and after some prodding—and a post by EFF—Sony Music agreed to withdraw its claim.

We're glad Sony stopped trying to block or monetize HVBA’s video. But the company’s response is troubling all the same. A Sony executive emailed HVBA to say that the company “has decided to withdraw its objection to the use of its two sound recordings” and “will waive Sony Music’s administrat[ive] fee.” That sounds like Sony was simply acting out of courtesy, when in fact the company had no right to demand a fee, by any name, for an obvious fair use. Other YouTube users with less knowledge of the law may have been convinced to pay Sony $500 or more, and provide detailed information, for uses of the music that the law makes free to all.

As Congress and the Copyright Office review the law and examine the effectiveness of automated systems like Content ID, they should keep in mind that automated flagging or filtering combined with misleading statements about a company’s legal rights can lead to abuse. That's another reason why YouTube-style automatic filtering should never be mandated by law, and why we need real penalties for false takedowns.