Formula 1 Digital Products Terms (NO DATA MINING)

[LucasCorsino]

Hi FastF1 maintainers and community—first of all, thank you for the amazing work you do. 🙌

Context: Formula 1 has updated its Digital Products Terms of Use to explicitly prohibit text/data mining and web scraping of its digital products (website, app, F1TV, etc.) and to withhold consent for using any data from those products to develop, train, fine-tune, or validate AI systems/models. The terms also include an express reservation of rights referencing Article 4(3) of the EU Digital Copyright Directive.

Why I’m raising this: Many users (myself included) use FastF1 for educational analysis and to produce non-commercial insights (lap times, stint pace, pit stops, strategy). I want to ensure that FastF1 users clearly understand what is and isn’t compliant under the updated F1 terms—especially regarding:

• whether any FastF1 functionality interacts with F1 “Digital Products” endpoints,

• the permitted sources for data (e.g., FIA public documents vs. data obtained from F1’s digital products),

• and the restrictions on redistribution of bulk/raw datasets or near-live timing.

Requests / suggestions:

1. Documentation note: Add a prominent README/Docs warning summarizing the new F1 terms and the potential contractual implications for users.
3. Source clarity: Document which modules/functions (if any) access endpoints that could be considered F1 “Digital Products,” and provide guidance on safe/alternative data sources (e.g., FIA public docs, properly licensed third-party providers).
5. Usage guidance: Recommend best practices for users (post-event analysis only, avoid bulk dataset redistribution, no logos/branding that imply affiliation, nominative references only, etc.).
7. AI use disclosure: Explicitly state that data from F1’s Digital Products must not be used for AI training/fine-tuning/validation where the F1 terms apply, and suggest “clean-room” workflows for users who need AI-related work with compliant sources.
9. FAQ & examples: Provide examples of compliant vs. risky workflows, plus a short “not legal advice” disclaimer so users can seek their own counsel if needed.

This issue is raised in good faith to help the community stay aligned with both the project’s Code of Conduct and the evolving IP/contract landscape. If helpful, I’m happy to draft a PR that adds a “Legal/Compliance Considerations” section to the docs with neutral, factual language and links to official terms.

Thanks again for all your work—and for any guidance you can provide here.

—
Not legal advice.

[theOehrly]

Hi, this is a very difficult topic in its sheer complexity. There are a bunch of rules, terms of use and guidelines that overlap here and that are very difficult to correctly interpret.

Relevant for this discussion are (at least):

• https://account.formula1.com/#/en/terms-of-use
• https://www.formula1.com/en/information/legal-notices.7egvZU48hzrypubGBNcQKt
• https://www.formula1.com/en/information/guidelines.4EOKE9RRqevL4niTK9kWyt

Importantly, Section 4 of the Guidelines makes everything rather muddy.

Followed by some more details and among other things

Regarding your questions/notes

• whether any FastF1 functionality interacts with F1 “Digital Products” endpoints,
• the permitted sources for data (e.g., FIA public documents vs. data obtained from F1’s digital products),

Consider all parts of FastF1 to interact with these endpoints except for the Ergast API interface (fastf1.ergast)

• and the restrictions on redistribution of bulk/raw datasets or near-live timing.

In general, both should be considered as not allowed except probably in some special cases that I am not going to define (and near-live timing is unsupported by FastF1 anyway).

Documentation note: Add a prominent README/Docs warning summarizing the new F1 terms and the potential contractual implications for users.
Source clarity: Document which modules/functions (if any) access endpoints that could be considered F1 “Digital Products,” and provide guidance on safe/alternative data sources (e.g., FIA public docs, properly licensed third-party providers).

A full overhaul of the documentation is planned and already in the works. I just need to get it done finally. I am planning on including this type of information there.

Usage guidance: Recommend best practices for users (post-event analysis only, avoid bulk dataset redistribution, no logos/branding that imply affiliation, nominative references only, etc.).

Maybe can be added to the documentation as well. It's difficult to get this right, I think, but I agree that it would be helpful for users.

AI use disclosure: Explicitly state that data from F1’s Digital Products must not be used for AI training/fine-tuning/validation where the F1 terms apply, and suggest “clean-room” workflows for users who need AI-related work with compliant sources.

What is AI in this context? (I know it's their wording.) Are the solely referring to LLMs and similar? I would assume that this is what they are targeting with this paragraph. And I know that Perplexity AI was developing very targeted F1 products, so this may even be a reaction to that. But what about a community project using machine learning for predicting results or pit stop strategies? The community terms may allow this again.

FAQ & examples: Provide examples of compliant vs. risky workflows, plus a short “not legal advice” disclaimer so users can seek their own counsel if needed.

This is getting into very difficult to get right territory. I think a lot of the community usage depends on F1's good will and their "endeavour to take a more relaxed approach when Fans engage in activities concerning FORMULA 1 Rights". And to some extent, I know for a fact that they are/were doing this. Therefore, I think such an example section is not very desirable, and I'd rather stick to a section with guidelines.

To summarize my opinion, for non-commercial community usage, the exact line of what is allowed is somewhat difficult to draw with only some things clearly not being allowed. Here, a section with guidelines and recommendations in the documentation could be very helpful.
For commercial usage, the rules are very strict. I don't really care about summarizing this. Everyone who intends commercial usage can pay for their own lawyer and for licensing if necessary. I don't want to get involved in any part of this.

If you want to create an informative section for this, your contribution is very welcome.

(Disclaimer: not legal advice)