ExTwitter Mostly Wins Silly Music Copyright Lawsuit

Last year, we wrote about a very silly lawsuit that some big music publishers had filed against ExTwitter, making some silly claims about how copyright law works. It basically ignored the existence of the DMCA, which was designed to prevent lawsuits like this one, where there is some infringement happening on the platform, but no realistic way for a website to police it, because it can’t know what is and is not infringing. That’s why the DMCA created a whole notice and takedown setup. This lawsuit seemed to basically ignore all that.

That’s part of the reason we thought that ExTwitter’s motion to dismiss was very strong and hoped it would carry the day. It turns out that it mostly did. A part of one claim survives, which ExTwitter should be able to get dismissed at the summary judgment stage if the company didn’t do something very, very stupid (which, these days, is no guarantee). But, on the whole, this is a good ruling for not just ExTwitter, but against dumb copyright lawsuits which have been back on the rise of late.

The ruling is pretty straightforward, dismissing most of the really silly claims from the publishers. The judge understood the basic DMCA issues, noting right up front that the DMCA creates a framework for sites to host user-generated content, without being held liable for infringement by users so long as they comply with the safe harbor requirements in the DMCA. But the music publishers are trying to get around that by claiming that ExTwitter can still be held liable by encouraging infringement.

As with so many copyright cases these days, there are different issues regarding whether or not the defendant engaged in direct infringement, contributory infringement, and/or vicarious infringement (which is often confused with contributory infringement, even by lawyers). Here the judge notes that the music publishers’ arguments on all three types of infringement are pretty flawed.

On direct infringement, it should be a very, very easy call, since it’s not ExTwitter itself that is uploading infringing videos. But the music publishers made use of the giant copyright mess created by the Supreme Court in the Aereo case a decade ago to argue that the Aereo “looks like a duck” test should magically apply here. They argued that ExTwitter was a type of broadcaster “transmitting” the infringing works, similar to Aereo’s rebroadcasting of TV channels. Thankfully, the judge is quick to see that the Aereo case was very, very different:

The plaintiffs argue that their theory of direct infringement falls squarely within Aereo, but it is hard to see how that could be the case. These plaintiffs’ allegations, in contrast with the bilateral transmission relationship at issue in Aereo, require the court to consider the respective roles of three parties: one who sent protected material, one who received it, and a third party, X Corp., that continuously operated the platform through which that infringement—and numerous other, non-infringing communications—occurred. X/Twitter, unlike Aereo, did not “transmit” any of the allegedly infringing material in the manner of a cable provider, because it was not the party that initially diverted that material from the intended channels of distribution. X/Twitter was more like a telephone company—providing the mechanism for communication between independent communicators—than like a cable company that actively selects material to make available. The purpose of the Transmit Clause, the Supreme Court acknowledged, was to “erase[] the . . . line between broadcaster and viewer,” but X/Twitter was neither of those things….

(Before anyone gets too hyped up about the comparison to telecom providers and phrases in here that sound common carrier-like, this is in a wholly different context — just looking at the copyright liability question, not everything else).

The Aereo majority, moreover, specifically acknowledged that “Congress, while intending the Transmit Clause to apply broadly to cable companies and their equivalents, did not intend to discourage or to control the emergence or use of different kinds of technologies.” Id. at 449. The Court stressed that it did not intend its “limited holding” to “have that effect.” Id. That unintended effect, however, is exactly what the plaintiffs would have this court endorse. There is no plausible case that X/Twitter is the “equivalent” of a cable company in the manner that Aereo—which had no meaningful existence other than as a copyright workaround for television broadcasts—was. The Aereo majority explained that it was not trying to “answer more precisely how the Transmit Clause or other provisions of the Copyright Act will apply to technologies not before” it, and that is how the court will construe its holding.

And thus, without anything clear linking ExTwitter to the direct transmission, the direct infringement falls:

As the Supreme Court explained in Aereo, the Transmit Clause was adopted with the specific purpose of ensuring that both the “broadcaster” and the “viewer” of an audiovisual work could, where appropriate, be held liable for direct infringement of the type involved in the tr.ansmission of broadcast television through cable systems. Aereo, 573 U.S. at 441. That purpose is consistent with the conclusion that “transmission” refers to the actions of the sender and/or ultimate recipient of a copyright-protected work—not those of the operators of the channels through which that transmission was accomplished. Claims against such a third party continue to be appropriate for consideration in connection with theories of secondary liability, not direct infringement. The court, accordingly, will dismiss Count I.

Next up is contributory infringement, which is where all the major action in copyright cases these days seems to live. Here, the judge is also not entirely impressed with the music publishers’ arguments, but does allow a small part of the claim to live on.

The concept of contributory infringement is a Supreme Court-invented concept, where they came up with an “inducement” standard found nowhere in copyright law itself, but which they felt was necessary to kill file sharing apps such as Grokster. In short, even if the app isn’t directly infringing, if it’s somehow taking proactive steps to encourage others to infringe, that can be seen as contributory infringement.

The problem here is that the music publishers claim that ExTwitter is engaging in contributory infringement just by existing and allowing people to upload music. If that were true, the entire DMCA notice-and-takedown procedure would be obsolete and dead. The judge is not generally impressed:

The plaintiffs, however, have not identified any caselaw or statutory law suggesting that merely hosting infringing content, until it is subject to a takedown notice, constitutes “materially contributing” to infringement under current law. Nor do the plaintiffs identify any authority that would support an argument that the operator of a social media platform materially contributes to infringement simply because there are some preventive steps that the operator could have taken but did not.

Even more to the point, the judge says setting up generally useful tools that might also be used for infringement is, in no way, indicative of contributory infringement:

Insofar as the plaintiffs are seeking to pursue a broad theory that X Corp. is liable for all of the infringement done on its platform because it has, in effect, created a straightforward, intentional infringement facilitation device, like the peer-to-peer filesharing applications that led to Grokster and similar litigation, the plaintiffs have failed to allege that theory in a manner consistent with either Grokster or the general law of secondary liability. Many of the supposedly problematic practices that the plaintiffs identify are unremarkable features of X/Twitter generally that X Corp. has simply failed to fence off completely from infringers. For example, while the plaintiffs make much of X/Twitter’s monetization of infringing tweets by surrounding them with paid-for promoted material, there is no allegation that those practices were meaningfully different than those X/Twitter applied to monetize popular, but entirely non-infringing, tweets. The plaintiffs complain that it is too easy to upload infringing audiovisual files onto X/Twitter, but, again, the plaintiffs are simply discussing a general feature of the platform. They do not allege that it is easier to post an infringing file than a non-infringing one. Any feature that makes a service easier for all of its users will, by definition, also make the service easier for bad actors. The plaintiffs have not identified any basis for concluding that X Corp. was obligated to make its service worse for everyone, just to punish the people who misuse it.

That final line is a good one.

That said, the judge does not dismiss this entire claim, leaving until later a few specific actions that ExTwitter is accused of, which might cause it problems if the company actually did some stupid stuff. If ExTwitter handled these situations in the way that most thoughtful user-generated content companies would handle it, with competent policies and lawyers, then ExTwitter won’t have a problem. Of course, this is Elon Musk’s company we’re talking about, so there’s no guarantee that they did the right thing. But we’re going to find out.

First up: did ExTwitter allow paying users to infringe more? If so, that could be trouble for the company:

The plaintiffs’ discrete allegations regarding some specific practices, however, much more plausibly fall into the category of materially contributing to infringement. Particularly striking is the allegation that X Corp. enforces its copyright policies less stringently against individuals willing to pay for its “verified” service. If X Corp. truly did allow some users to effectively purchase the right to be able to infringe with less severe consequences, then that was plausibly an instance of “promoting” X/Twitter’s “use to infringe copyright, as shown by . . . affirmative steps taken to foster infringement,” which Grokster acknowledged as a sufficient basis for liability.

This is the kind of thing that a well-run company would have documentation on to show that it’s just not true. Hopefully that’s the case with ExTwitter. Assuming the company still has competent policies on this stuff in place, they should be able to demonstrate that during discovery. If they do, however, treat paying customers differently for handling copyright takedown than… holy shit would that be stupid. I guess we’ll find out.

Second: was the company deliberately delaying responses to DMCA notices? This one’s trickier, because there’s no defined period of time in which a website has to respond, but anything that indicates they slow walked things could be trouble.

Similarly, if X Corp. engaged in egregious delays in responding to valid takedown notices, or outright ignored some notices that were both facially and actually valid, that could support liability. The plaintiffs have not identified any basis for concluding that X Corp. had an obligation to respond to notices of infringement either blindly or instantaneously. In fact, a company that instantly complied with every takedown notice filed, without scrutinizing it at all, would run the risk of enabling abusive, anticompetitive takedown practices—a danger that the DMCA itself acknowledges. See 17 U.S.C. § 512(f) (creating liability for some abusive takedown practices). The fact that some delay may be appropriate, however, is not an unlimited license to drag one’s feet inordinately or forever. If, in fact, X Corp. allowed delays to extend beyond what was reasonably necessary to process takedown requests, in order to make the platform a more attractive tool to infringers, that improper extension of delays would plausibly amount to “purposeful, culpable . . . conduct” intended to enable infringement, capable of supporting secondary liability.

Again, a well-run company would have the documentation necessary to debunk these claims. Does ExTwitter? ¯\_(ツ)_/¯ We’ll find out.

Finally, in the contributory infringement space, there’s the question of whether or not ExTwitter has a functioning repeat infringer policy. Again, if the company does (and it used to…), it should be able to cough up evidence to that effect:

Finally, the plaintiffs have plausibly alleged that X Corp. engaged in contributory infringement by failing to take meaningful steps to address the actions of severe serial infringers. Nothing in the Complaint is sufficient to plausibly suggest that a social media platform like X/Twitter has an obligation to suspend or terminate the account of every person who infringes more than once, or even every user who infringes a number of times. Like delays, however, recidivism can exist in degrees. The plaintiffs have alleged that there was an identifiable subset of X/Twitter users who openly and obviously used the service as a tool for repeatedly posting infringing content, but X Corp. affirmatively declined to take reasonable steps in response to those users’ actions. Again, there is no basis in the law for concluding that the operator of a social media platform will face liability simply because it was less draconian in its enforcement than copyright holders would prefer. If, however, there was a class of X/Twitter users who were brazenly using the platform as an infringement tool, and X Corp. made the decision to unreasonably withhold enforcement of its own policies against those users, with the foreseeable consequence of ongoing infringement, then X Corp. could plausibly be held contributorily liable.

So those bits of possible contributory infringement live on. This means that there will likely be discovery on those issues. If ExTwitter produces evidence that the claims by the publishers are hogwash and they do handle all those things appropriately, the company will likely move for summary judgment and win. If the company under Musk has done something very, very stupid regarding copyright enforcement, then… the company would probably be wise to settle and get the case off the books.

As for the vicarious infringement claim, it’s amazing to me how copyright holders always claim vicarious infringement, and we almost always find out it’s not vicarious infringement. It’s becoming the RICO of secondary infringement claims.

To vicariously infringe, it has to be shown that the website is directly profiting off of the infringement (not just general usage) and, because of that, fails to take steps to prevent it. But you can’t just find some infringement on a platform and say it’s vicarious. However, that’s more or less what the publishers did here. The court is not impressed.

The Supreme Court has recognized, however, that the Copyright Act does not require the seller of a useful, lawful product to scrap its entire business just to spite infringers. The Court faced that very dilemma in connection with the rise of videocassette recorders—that is, VCRs—in 1994, and it held that vicarious liability does not arise merely because a company “sold [a product] with constructive knowledge of the fact that [its] customers may use that to make unauthorized copies of copyrighted material.” Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. at 439. When the caselaw involving vicarious infringement discusses a defendant’s “right to stop or limit” infringement, then, it means something more than simply the right to refuse to distribute one’s product to possible infringers.

The court notes that most vicarious infringement situations involve cases where an employee or an “agent” of the company is engaging in infringement to the benefit of the company. And there’s no evidence of that here. Indeed, the court basically says that the publishers are sort of pleading contributory infringement factors and claiming they’re also vicarious. But they’re not. So those claims are tossed as well:

X Corp. undoubtedly had some power over X/Twitter’s users—the way that a company that provides a valued service always has power over the customers who rely on it—but that does not turn customers into even loose equivalents of agents or subordinates. See Music Force, LLC v. Sony Music Holdings Inc., No. CV 19-6430 FMO (RAOx), 2020 WL 5733258, at *3 (C.D. Cal. Aug. 12, 2020) (“[T]he right to terminate services or a contract with an infringer does not amount ‘to a right and ability to supervise the infringing conduct.’”) (quoting Routt v. Amazon.com, Inc., 584 F. App’x 713, 715 (9th Cir. 2014)). As with the issue of direct infringement, the plaintiffs are trying to force X Corp’s actions into a category not intended to account for the actual character of the relationships at issue, when there is a tool—the doctrine of contributory infringement— uniquely suited to the job.

And that’s it. Basically all three theories by the publishers are flawed. A few small parts of the contributory claims live on, but should be easily disposed of as long as Elon didn’t do something stupid.