Meta business terms are the rules Meta sets for commercial accounts on Facebook, Instagram, Messenger, Threads and WhatsApp. If you run a page, profile or channel through a business account, they apply to you. The Copyright Alliance says that covers virtually every organisation in the creative industry. Its members include the RIAA, the NMPA and Universal Music Group, and it has now called the new version of the Meta business terms “unconscionable”. A day later, Meta said the November 20 rollout could slip again. Below you’ll find the full timeline, what the draft says, and what to check if your music lives on Meta.
Most people accept terms of service the way they accept cookie banners: one thumb, mid-scroll, thinking about lunch. This time a lot of people actually read them.
Meta business terms: what happened so far
Music Business Worldwide has the most complete version of the story. Here are the dates:
- October 15: Nieman Lab first reports on the new Meta business terms. An early draft says Meta “may use any content from such domains, web pages, or URLs to provide, develop, improve, research, train, and maintain the integrity of” its products.
- Later in October: publishers push back, so Meta drops the word “train”.
- October 23: Meta moves the effective date from October 30 to November 20. Senator Marsha Blackburn, a Tennessee Republican, asks Meta to withdraw the terms.
- The next day, Senator Adam Schiff, a California Democrat, writes to Mark Zuckerberg and points out that the terms never mention robots.txt.
- October 27: the Copyright Alliance publishes its statement.
- One day later, Andy Stone, Meta’s VP of Communications, says on X that implementation will wait if the consultation runs past November 20.
Reuters reported the rollout as delayed indefinitely. On the same day, according to AFP, a court in Oslo temporarily barred Meta from applying its new business terms after ten Norwegian media groups, including NRK and Schibsted, challenged them. The court acted on an emergency basis, without hearing from Meta. Its order holds until a final ruling on the terms’ legality.
Meta’s website still showed November 20, Engadget noted. So the date is official and probably fictional. Both at once.
What the draft says about your website
According to Engadget’s quote, the current draft says Meta may “use automated software (such as web scrapers, bots, robots, spiders, crawlers, and user agents)” to access business websites “in order to provide, develop, improve, research, and maintain advertising and business products”. The word “train” is gone. “Business products” stayed. Nobody outside Meta knows where that phrase ends.
In plain words, the Meta business terms would let the company’s bots read your website. For a label or a composer, that site holds the catalogue, the lyrics, the press photos and the liner notes. Everything, basically.
The Copyright Alliance statement says the changes “appear to authorize Meta to scrape and use websites and other content from commercial account holders for Meta’s business products, including to train and ground Meta’s AI products”. To “ground” an AI means feeding it live content so its answers draw on that content. There is no opt-out, the statement adds. So a company that disagrees with the Meta business terms can accept them or close the business accounts it uses to reach its audience. The Alliance calls that second option “extremely difficult given Meta’s market power”.
Robots.txt and the new Meta business terms
Meta disputes this reading. Asked about Blackburn’s letter, the company told Reuters that “content owners will still be able to use robots.txt to signal preferences” and that “this isn’t about scraping news sites.”
A quick tangent about robots.txt. It’s a small text file at the root of a website that tells crawlers which pages they may visit. Its legal force is roughly that of a “please knock” note taped to an open door. Legal experts quoted by Engadget say it isn’t binding and that AI companies regularly ignore it.
Anyway. The Alliance says the Meta business terms let the company “ignore or circumvent” robots.txt. Schiff’s complaint is shorter. The contract doesn’t mention it at all.
Why music groups care about Meta business terms
News publishers raised the alarm first, Music Ally reports. But the Copyright Alliance says this goes far beyond newsrooms, and its member list makes the case: A2IM, ASCAP, BMI, SESAC, SoundExchange and the Recording Academy, on top of the RIAA, NMPA and UMG. That’s a lot of acronyms. The group says it represents more than two million individual creators and more than 15,000 organisations in the US.
Universal is the interesting member. UMG renewed its licensing deal with Meta in August 2024, and at the time both companies said they would work together “to address, among other things, unauthorized AI-generated content that could affect artists and songwriters.” Now a trade group UMG belongs to calls Meta’s contract unconscionable. Awkward? Less than it looks. The licensing deal covers the music people put in Reels, while the Meta business terms cover what its crawlers may do on your site.
Then there’s the lawsuit. Wixen Music Publishing sued Meta in January, alleging that its “reason for slashing payments to songwriters is to replace human-generated, royalty-bearing music with royalty-free AI-generated music.” In September a federal judge threw out Wixen’s copyright claims on standing, so the ruling never decided if Meta infringed. Wixen filed a second amended complaint on September 18.
The terms you click are turning into AI licences
Creators should underline this part. In June, Google moved to dismiss a lawsuit brought by independent musicians over its Lyria 3 music model, MBW reports. Its argument: the artists had licensed their music when they accepted YouTube’s terms of service. The musicians say Google trained Lyria 3 on their recordings without permission or payment.
The pattern is easy to see. When an AI training fight reaches court, the platform points to the terms you clicked years ago. The Meta business terms would put the permission to crawl in writing before any fight even starts. If a case gets past that argument, it usually lands on fair use, which is where the Anthropic lyrics case now stands.
Rights on social platforms were messy long before anyone’s bots showed up. Mubert’s own data, last 28 days: about one in four results from our copyright checker found a rights holder’s claim on the uploaded track. That share fell slightly, from 26.4% to 25.7%.
What we think
“Unconscionable” is a legal word, not a mood. It describes a contract so one-sided that a court may refuse to enforce it. Lawyers reach for it when “unfair” feels too polite.
On the narrow point, the Copyright Alliance is right. These business terms would let Meta read your website for undefined “business products”, with no opt-out and no mention of robots.txt. They ask for trust Meta hasn’t put in writing. If the Alliance’s reading is wrong, the fix is one sentence in the contract. Not a post on X.
The delay is good news. It isn’t a win. Stone’s post moves a date and changes no wording, and since the existing terms stay in place for now, nobody has been promised anything about the next draft. (We have a Facebook page too. We read this draft twice, which is two more times than we read the last one.)
What to do before the new Meta business terms land
If your Facebook page also earns money, the monetization rules are a separate maze. For the contract itself:
- Check if your page or profile runs through a Meta business account, since those are the accounts the new terms cover.
- Read the next draft on Meta’s own site, not the notification summary.
- Add robots.txt rules for crawlers you don’t want, knowing they’re a request and not a lock.
- Move stems, unreleased audio and full lyrics behind a login instead of leaving them on open pages.
- Ask your label group, PRO or trade body what it is sending to Meta’s consultation.
- Watch for a new effective date, because the old terms apply until the consultation ends.
Cookie-banner rules still work for everything else online. Read this one with both thumbs.
Who we are
