In a blockbuster ruling that is bound to have far-reaching implications, including in the swarm of copyright infringement cases brought against AI platforms, a California federal court recently dismissed a complaint that X Corp. (f/k/a Twitter) brought against web scraping company Bright Data.  See X Corp. v. Bright Data Ltd., No. 23 Civ. 3698 (N.D. Cal. May 9, 2024).  Judge Alsup held, among other things, that the Copyright Act preempted the social media giant’s breach of contract and other state law claims because their enforcement would conflict with federal copyright law.  Below we cover the highlights, including the court’s groundbreaking application of conflict-based copyright preemption to a contract claim.

X Sues Bright Data For Web Scraping

X alleged that Bright Data scraped and sold X’s data, using complex tools to circumvent X’s anti-scraping technology, while inducing Bright Data’s users to participate in the scraping—all in violation of X’s terms.  X asserted claims for breach of contract, tortious interference, unjust enrichment, trespass to chattels, violations of California Business and Professions Code Section 17200, and misappropriation.  The court distilled X’s complaint into “two grievances”: that Bright Data had improperly (1) accessed X’s systems, and (2) scraped and sold X’s data. 

Court Dismisses Claims Based on Improper Access

Trespass to Chattels.  With respect to the first “access” category, the court started with “trespass to chattels, an old tort that is finding new life.”  The court held that the claim failed because X did not adequately allege that Bright Data’s access to X’s systems damaged X through “impairment or deprivation.”  The court was not convinced that “sending requests to X Corp.’s servers with a scraper is inherently burdensome, or inherently more burdensome than an X user sending requests to X Corp.’s servers with a browser.”

Violation of Section 17200.  X claimed violations of the unlawful and fraudulent prongs of California’s “catchall” unfair competition law, Cal. Bus. Prof. Code § 17200.  The court held that X failed to allege any fraudulent business act by Bright Data, rejecting X’s argument that Bright Data meaningfully deceived X by “posing” as legitimate users or deploying widely available IP proxy technology.  With respect to the unlawful prong, the court held that X failed to state any predicate claim as needed to establish a violation.

Tortious Interference.  X alleged that Bright Data induced its users to breach their agreements with X by providing them with automated tools to access and scrape data from X’s servers. As to access, the court held that X had failed to allege a viable theory of damages, for the same reasons explained in the trespass to chattels section.  The court addressed the alleged scraping breach in the scraping section of its opinion.

Breach of Contract.  X claimed that Bright Data directly breached its contract with X by accessing its platform through unauthorized means and scraping and selling X’s data.  The court again found the access-based breach deficient, holding that X had “not alleged any damage resulting from access through unauthorized means, merely alluding to diminished capacity and reputational harm without pleading any impairment or deprivation of servers.”  The court addressed the scraping and selling component of the breach claim in the next section.

Court Dismisses Claims Based on Improper Scraping and Selling of Data

The court began its analysis of the data scraping/selling claims by observing that, whereas X owns the technology that provides the basis for X’s access-based claims, X’s users hold the rights in the content that provides the basis for X’s scraping/selling claims.  The court further observed that X’s users grant X only a non-exclusive license to use that data; yet, X was seeking to exclude others (i.e., Bright Data and its customers) from using that content, which is not a privilege the Copyright Act extends to non-exclusive licensees.  “So how does X Corp. purport to do this?”  Through its terms, which prohibit scraping.  “To the extent X Corp.’s claims are based on scraping and selling of data, they rest on these contractual provisions.”

Enter Copyright Preemption

With this observation, the court turned to copyright preemption, beginning with the “upshot” of its opinion:

The upshot is that, invoking state contract and tort law, X Corp. would entrench its own private copyright system that rivals, even conflicts with, the actual copyright system enacted by Congress. X Corp. would yank into its private domain and hold for sale information open to all, exercising a copyright owner’s right to exclude where it has no such right.

The court then reviewed the contours of federal and copyright preemption, noting that federal law may preempt state-based claims both expressly, by statute, and impliedly, when the state law conflicts with federal law.  While “conflict preemption has played second fiddle to express preemption in the caselaw as of late,” the court opined that “it is the more appropriate consideration when the question presented is not whether rights created by state law are equivalent to rights created by federal copyright law but whether enforcement of state law undermines federal copyright law.” 

The Risk of “Information Monopolies”

After citing the Supreme Court’s standard for conflict preemption (where enforcement of a state law “stands as an obstacle” to federal law), Judge Alsup recalled the Ninth Circuit’s salient observation in another important web scraping case:

We agree with the district court that giving companies like LinkedIn free rein to decide, on any basis, who can collect and use data—data that the companies do not own, that they otherwise make publicly available to viewers, and that the companies themselves collect and use—risks the possible creation of information monopolies that would disserve the public interest.

hiQ Labs, Inc. v. LinkedIn Corp., 31 F.4th 1180, 1202 (9th Cir. 2022).

In re Jackson: The Second Circuit’s Conflict-Based Copyright Preemption Test

Since neither hiQ nor any other Ninth Circuit case had ruled on the question presented, the court turned to “persuasive authorities”—namely, the Second Circuit’s thoughtful analysis of conflict-based copyright preemption in In re Jackson, 972 F.3d 25 (2d Cir. 2020). 

In In re Jackson, Judge Leval articulated a two-part test for determining whether the application of state law would “interfere with or frustrate the functioning of the regime created by the Copyright Act.”  Courts should examine whether the claim (1) conflicts with federal copyright policies; and (2) asserts a substantial state interest, distinct from the interests underlying federal copyright law. In In re Jackson, the Second Circuit applied this test to find that a rapper’s claim for right of publicity was impliedly preempted by the Copyright Act. 

Citing In re Jackson, the court concluded that “the extent to which public data may be freely copied from social media platforms, even under the banner of scraping, should generally be governed by the Copyright Act, not by conflicting, ubiquitous terms.” 

X’s Anti-Scraping Terms Conflict with Copyright Law 

The court found three ways that X’s “state-law claims based on scraping and selling of data undermine the purpose and intended effects of the Copyright Act.” 

First, X’s scraping claims would enable X, a non-exclusive licensee, to block others from exercising rights that, under the Copyright Act, belong exclusively to X’s users. 

Second, enforcing X’s contractual scraping prohibitions would frustrate fair use by restricting the rights that users otherwise would have to make fair use of copyright-protected works without permission or payment. 

Third, the anti-scraping provisions purported to give X “de facto copyright ownership over content that Congress declined to extend copyright protection to in the first place (e.g., likes, user names, short comments),” thereby shrinking the public domain in contravention of key copyright policy.

X’s Anti-Scraping Terms Seek to Protect Interests in the Sphere of Copyright

After identifying these conflicts, the court proceeded to the second part of the In re Jackson test, which examines “whether a state-created right vindicates a substantial state law interest, i.e., an interest outside the sphere of congressional concern in the copyright] laws, that is distinct from the interests served by the federal law which may preempt the claims” (cleaned up). The court gave “users’ privacy” as an example of a state-law interest that may be distinct from the interests of copyright. 

The court determined that X was not seeking to vindicate privacy or any other non-copyright-related interest. Rather, X wanted the sole right to monetize its data.  “X Corp. is happy to allow the extraction and copying of X users’ content so long as it gets paid.”  The court found that X’s scraping claims “amount[ed] to little more than camouflage for an attempt to exercise control over the exploitation of a copyright” (quoting In re Jackson, 972 F.3d at 38). 

Thus, the claims based on the scraping and sale of data were held to be preempted.

Some Thoughts About the Case

Here’s a few thoughts about the significance of X Corp. v. Bright Data.

  • This case represents the first time that a district court has applied In re Jackson’s conflict preemption test, but it likely won’t be the last. As conflict-based copyright preemption gains traction, the variety of situations in which the doctrine is held to apply will expand, offering a powerful and malleable line of defense against contract and state law claims involving interests within the sphere of copyright.
  • While prior decisions have held that contract claims were expressly preempted under Section 301 of the Copyright Act, this is the first case in which a contract claim was dismissed under principles of conflict preemption.  This too may be the start of a trend.  As observed in a law review article cited in both X Corp. and In re Jackson, conflict preemption “might be the primary and, in many cases, the only question that should be asked” when examining contract claims.  Guy A. Rub, A Less-Formalistic Copyright Preemption, 24 J. Intell. Prop. L. 327, 345-6 (2017).
  • It is important to understand that, unlike express/statutory preemption, conflict preemption does not require “equivalence” between the state law claim and a copyright claim.  Thus, a claim that seeks to enforce a contractual provision that prohibits activities beyond reproduction, distribution, etc. may still be preempted if enforcement of the provision would stand as an obstacle to the federal copyright regime. 
  • The court noted that the case before it concerned “a massive regime of adhesive terms imposed by X Corp. that stands to fundamentally alter the rights and privileges of the world at large (or at least hundreds of millions of alleged X users).”  A different outcome might result if the case concerned “an arm’s length contract between two sophisticated parties in which one or the other adjusts their rights and privileges under federal copyright law.”
  • As we previously blogged, copyright preemption is a key issue in the infringement cases brought against AI platforms.  This case drives the point home, especially the court’s sweeping opinion that “the extent to which public data may be freely copied from social media platforms, even under the banner of scraping, should generally be governed by the Copyright Act, not by conflicting, ubiquitous terms.”