YouTube video recognition

This article was originally on a blog post platform and may be missing photos, graphics or links. See About archive blog posts.

Months after promising to do so, Google has rolled out a recognition system for YouTube that can help copyright owners block unauthorized uploads of their work. Gigi Sohn of Public Knowledge weighed in with an early criticism, noting that such systems have trouble distinguishing fair uses from infringements. So do some copyright holders, heh heh heh.

In fairness to Google, content ID systems can be calibrated to reduce the chance of blocking mash-ups, reviews and other uses of copyrighted material that fit under the ‘fair use’ roof. But it’s hard to build an advertiser-supported business model without the ability to tell advertisers and content owners what’s being shown, how often and to whom. That’s where content ID tools could make a real contribution. And that’s what the tool makers have been saying repeatedly in recent months: the best use of the technology is in helping to monetize content that’s being made widely available, rather than trying to stop people from uploading clips that they don’t own.

YouTube is hardly the only site deploying these tools -- other active testers include MySpace, Vuze and Veoh. Ultimately, the most influential voice in how content recognition is used won’t be Hollywood’s or YouTube’s, but the courts’ -- particularly if Google and Viacom don’t strike a deal before the latter’s lawsuit against YouTube goes to trial.