Two recent decisions from the Northern District of California break new ground in the interpretation of the Digital Millennium Copyright Act (DMCA).  These cases may indicate how a couple of the most contentious issues about the use of copyrighted works in user-generated content on the Internet will be resolved.

Io Group, Inc. v. Veoh Networks, Inc., 2008 WL 4065872 [Case No. C:06-03926] (N.D. Cal. 2008), makes a number of important holdings in concluding that an online video sharing site (not dissimilar from YouTube) was protected from liability for copyright damages by a DMCA safe harbor.  Although still, at this point, an unpublished district court decision, it could become one of the most significant cases construing the DMCA safe harbors.  Lenz v. Universal Music Corp., et al., 2008 WL 3884333 [Case No. C:07-3782] (N.D. Cal. 2008), is the first case to squarely hold that copyright owners must conduct a fair use analysis before sending takedown notices pursuant to the DMCA, or face potential liability to users whose materials are the subject of such notices.


Io Group, Inc. v. Veoh Networks, Inc.


The Veoh decision arose from a dispute between an adult entertainment provider and Veoh, an online video sharing site.  Io Group sued after it discovered that portions of several of its copyrighted movies had been posted on the Veoh site by users.  Notably, it did not send notices to Veoh to remove the infringing material before suing. 


The district court’s decision―granting Veoh’s motion for summary judgment and denying Io Group’s―provides a detailed analysis of the DMCA safe harbor for infringement resulting from material stored on an Internet service provider’s system “at the direction of a user.”  (17. U.S.C. 512(c).)  Presuming infringement, the decision focuses exclusively on whether Veoh qualified for the protection from damages for copyright infringement provided by this safe harbor.


At the risk of over-simplification, Veoh uses an automated system to make video uploaded by partners and users available on its website.  The system converts uploaded video files into a digital format compatible with most Internet browsers.  In addition, it generates low-resolution stills or “screencaps” that are taken from the video, and may also create a video preview.  Videos are indexed using information (title, genre, etc.) provided by the user who uploads the video.  When a user searches for videos on the Veoh site, the index information and s screencap are provided in the search results.  In addition, the search results may include a link to additional screencaps, and a link to use to download the entire video. 


The district court first addressed the threshold requirements that online service provides must meet in order to qualify for any of the DMCA safe harbors.  It was not disputed that Veoh was a service provider, that it had adopted and informed users of a policy of terminating repeat infringers, and that it did not interfere with standard copyright protection measures.  The issue raised by Io Group and addressed by the court was whether Veoh reasonably implemented its termination policy.


The court held that it did.  In doing so, the court held that it is not necessary to track terminated users by some identification method (such as using their IP addresses) and block them from using the site, in order to prevent them from using a different Internet identity to post infringing material on the Veoh site.


Turning to the specific requirements of the safe harbor for material that resides on a website “at the direction of a user,” the court reached several important conclusions.  In order to understand the significance of these conclusions, it’s important to have in mind what the requirements of this safe harbor are. 


The Veoh decision provides a succinct description of those requirements.  At the outset, it must be shown that the alleged in infringement is occurring “by reason of the storage at the direction of a user of material that resides on a system or network controlled or operated by or for the service provider.”  In addition:


In essence, a service provider is eligible for safe harbor under section 512(c) if it (1) does not know of infringement; or (2) acts expeditiously to remove or disable access to the material when it (a) has actual knowledge, (b) is aware of facts or circumstances from which infringing activity is apparent, or (c) has received DMCA-compliant notice; and (3) either does not have the right and ability to control the infringing activity, or-if it does-that it does not receive a financial benefit directly attributable to the infringing activity.


Addressing the threshold requirement that the material at issue must be stored on the system “at the direction of a user,” the court made two significant points:  (1) The fact that Veoh’s system automatically converted the digital format of video files did not deprive it of protection; and (2) neither did the fact that the system automatically generated and displayed “screencaps” taken from those files.  The court noted that, unlike another DMCA safe harbor (17 U.S.C. § 512(a)), this safe harbor was not restricted to service providers who do not modify the content of material sent or received.  Hence, it said, the safe harbor is “not limited to merely storing material.”


The significance of this holding remains to be seen, but it suggests that the creation and display of derivative works though the use of automated systems will not deprive service providers of protection under this safe harbor.  In addition, it indicates that the safe harbor will be “format neutral.” 


Next, the court considered whether Veoh had actual or constructive knowledge of the infringement, and failed to promptly remove the infringing material.  It was undisputed that Veoh was not notified of Io Group’s materials, and hence did not have actual knowledge.  However, Io Group asserted that Veoh was aware of circumstances that made the infringement apparent. 


Addressing this claim, the court first adopted a very stringent standard for finding such constructive knowledge.  Following an earlier district court decision,[1]  the court said that such knowledge will be imputed only if a service provider “deliberatively proceeded in the face of blatant factors of which it was aware.”  This standard, which essentially imposes a requirement of either actual knowledge or reckless disregard of infringement, is familiar in the context of cases seeking to impose liability for speech protected by the First Amendment.  While not insurmountable, it represents one of the most stringent standards of proof in civil cases.  It would be very difficult for copyright owners to meet this standard in the absence of having sent takedown notices.


Next, the court addressed the requirement that the service provider either not have the right and ability to control the infringing activity, or that it not receive a direct financial benefit as the result of that activity.  The court did not reach the second question, because it found that Veoh did not have the ability to control the infringing conduct.  In doing so, it made two significant holdings―holdings that may not be entirely consistent, or which at least appear to provide alternate and independent bases for its decision. 


First, the court appears to conclude that only the ability to prevent the initial uploading of infringing material will be sufficient to constitute the “right and ability to control” infringing activity.  At a minimum, the court explicitly says that prescreening of all material prior to publication is not required.  Although arguably obvious, this conclusion is also fundamental to the operations of numerous online services based on the sharing of user generated content.  The burden of pre-screening would make the business models of many such services untenable.  The court recognized as much:  “[Io Group’s] not-so-subtle suggestion is that, if Veoh cannot prevent infringement from occurring, then it should not be allowed to exist.”  The court rejected that result.


Second, the court appears to hold that in order to be deemed to have the ability to control infringing activity, a service provider must be able to identify infringing material without receiving notice from the copyright owner.  Again, while this holding may seem self-apparent, it may be quite important.  If broadly applied, it could mean that service providers will be protected by the safe harbor unless they receive takedown notices or ignore blatant and unmistakable infringement.  In other words, in the absence of actual notice or constructive knowledge of infringement (pursuant to the stringent standard discussed above), a service provider could not be found to have the right or ability to control the infringement.


Finally, one other aspect of this decision should be mentioned.  While the DMCA does not expressly impose a requirement that service providers “police” their sites to prevent infringement, such policing may help them in qualifying for the DMCA safe harbors.  Veoh had and followed policies of promptly removing infringing material, terminating repeat infringers, preventing material identified as infringing from being re-posted, and even removing infringing material from user’s hard drives.[2]  Not all of these measures are required by the DMCA or the cases construing it, but they clearly influenced the court’s decision.


Because of the parallels between Veoh and YouTube, several commentators have suggested that the Veoh decision bodes well for YouTube in the litigation brought against it by Viacom and others.  It probably won’t have any direct impact, unless it is appealed and affirmed by the Ninth Circuit before a final decision in the YouTube litigation.  (Even then, there are a number of factual distinctions between the cases, and the Second Circuit, where the YouTube case is pending, could reach a different conclusion.)  However, it holds that the DMCA safe harbors are quite robust, and that is surely welcome news to online service providers with business models built on sharing user generated content, such as YouTube and many others.


Lenz v. Universal Music Corp.


The Lenz decision presents a single issue: must copyright owners conduct a fair use analysis before sending a notice to have material taken off an Internet web site in order to avoid potential liability under the DMCA for sending a takedown notice without a “good faith belief” that the material is infringing?  (See 17 U.S.C. § 512(c)(3)(A).)  The district court in this held that they must.


Lenz made a now notorious 29-second video of her two small children dancing to the indistinctly heard music of Prince’s “Let’s Go Crazy.”  She posted the video on YouTube.  Universal, which owns the copyright to the song, sent a takedown notice, and YouTube removed the video.  Lenz, sent a counter-notice and subsequently, represented by the Electronic Frontier Foundation, brought an action asserting that Universal had violated the DMCA by sending the takedown notice in bad faith.  (17 U.S. C. § 512(f).)[3]  The claim was premised on the assertions that the video was protected as fair use, and that Universal had failed to engage in any fair use analysis prior to sending the takedown notice.


Universal moved to dismiss the complaint.  It did not assert that it had engaged in a fair use analysis, but it contended that copyright owners are not required by the DMCA to engage in such an analysis prior to sending a takedown notice, and that if they are such an analysis need only be performed after a counter-notice is received.


The district court disagreed.  While expressing skepticism that Lenz would ultimately be able to prove the subject bad faith required to support her claim of misrepresentation,[4] the court held that fair use is a use “authorized by law” pursuant to the DMCA notice provision (17 U.S.C. § 512(c)(3)(A)).  Therefore, it concluded, copyright owners must engage in an analysis of whether the disputed material is protected as fair use before sending a takedown notice.


The question of fair use is a complex one.  It has been the subject of extensive and contentious debate between fair use advocates and copyright holders in the context of user generated content posted to websites such as YouTube.  The impact of this decision is difficult to gauge, given the very high standard for imposing liability on copyright holders under the DMCA (at least in the Ninth Circuit).  However, by imposing a requirement of at least a minimal fair use analysis before sending a takedown notice, this decision may alter the terms of the debate.


Authored by:

James M. Chadwick

(650) 815-2605


[1]           Corbis Corp. v., 351 F.Supp.2d 1090, 1108 (W.D. Wa. 2004).

[2]           Some caution with respect to this last measure may be warranted, at least unless notice of the claim of infringement has been provided to the user, and until the time for a DMCA counter-notice has run.

[3]           Lenz’s original complaint asserted claims for tortious interference with contract, violation of the DMCA, and declaratory judgment of noninfringement.  That complaint was dismissed with leave to amend, and the amended complaint apparently asserted a claim only pursuant to the DMCA. 

[4]           See Rossi v. Motion Picture Ass’n of America, Inc., 391 F.3d 1000, 1004 (9th Cir. 2004).