Meta Platforms v. Bright Data is a data-scraping case in which Meta, which operates Facebook and Instagram, sued Bright Data Ltd. for breach of contract and tortious interference. Meta said Bright Data had violated the Facebook Terms of Service and Instagram Terms of Use by scraping and selling data from the two services. On 2024-01-23 Judge Edward Chen denied Meta’s motion for partial summary judgment and granted Bright Data’s motion on the contract claim.[1]
Background
Bright Data sells proxy network services, scraping tools and datasets, including a Facebook Scraper and an Instagram Scraper. It said it collected only public data that a visitor can see without logging in.[1] Bright Data admitted that it held Facebook and Instagram accounts until 4 December 2022 and was bound by the terms while it did, but said it had then terminated those accounts and rejected the terms.[1]
The Facebook Terms said users may not “access or collect data from our Products using automated means (without our prior permission)” and may not “sell, license, or purchase any data obtained from us or our services”. The Instagram Terms contained similar prohibitions.[1]
The dispute
Meta demanded in November 2022 that Bright Data stop scraping and selling Meta user data, and filed suit in January 2023.[1] Both sides moved for summary judgment on the contract claim. Meta argued that Bright Data was bound by the terms and had breached them by scraping and selling data. Bright Data argued that the terms govern the “use” of the services by a user, and do not reach logged-off access to public pages.[1]
Decision or outcome
The court found that Meta had not shown that Bright Data scraped non-public data while logged in. It distinguished defeating anti-bot measures such as CAPTCHAs from entering a password-protected area, citing its earlier hiQ Labs v. LinkedIn decision.[1]
Interpreting the terms under California law, the court held that they apply to a “user” who “uses” the services, and that logged-off scraping of public data after Bright Data had ended its accounts was not covered. For Facebook it considered the survival clause in section 4.2, which keeps section 3 in place after an account is deleted, and read it as not extending to that conduct. The Instagram Terms had no survival clause, so nothing barred post-termination scraping or sale.[1] The court noted that applying the survival clause as Meta construed it would raise serious questions under the Restatement (Second) of Contracts, but did not decide unconscionability.[1] It allowed Meta three weeks to add a quasi-contract claim and noted that only the tortious interference claim otherwise remained.[1]
Significance for software licensing and SAM practice
The ruling is a district court decision on one set of platform terms. It shows that the reach of a terms-of-use restriction depends on who is a “user”, on what counts as “use” and on whether the terms survive termination. The same questions arise for clickwrap and browsewrap terms attached to software and data services, including generative AI licensing. A companion dispute over X’s terms is described in X Corp. v. Bright Data.
Lessons learned
- Terms bind users. The court read the prohibitions as applying to use of the service, so conduct by a logged-off non-user was outside them.[1]
- Survival clauses need limits and clarity. An open-ended survival clause was read narrowly, and a clause that does not mention survival did not carry the prohibition past termination.[1]
- Evidence of logged-in access matters. Meta could not show collection of non-public data, and the court did not treat circumventing CAPTCHAs on public pages as equivalent to passing a password gate.[1]
- Check which claims remain. The ruling disposed of the contract claim only; a tortious interference claim was still pending at the time of the order.[1]