Latest / Ctrl AI Profit / Ep. 257 | Meta's New Terms Let It Crawl Your Business Website Starting October 30
Transcript
- 0:20SPEAKER_01: If your business has a Facebook page or runs ads on Meta, new terms kick in on October 30th that say Meta can send robots to read the website tied to your business.
- 0:30SPEAKER_00: That's the news in one sentence.
- 0:32SPEAKER_00: Meta posted it in its business help center.
- 0:35SPEAKER_00: Updated versions of the meta commercial terms and the self-serve ad terms take effect October 30th, 2026.
- 0:43SPEAKER_01: Okay.
- 0:44SPEAKER_01: Commercial terms.
- 0:45SPEAKER_01: Who does that actually cover?
- 0:47SPEAKER_01: Because I don't think of myself as a meta business customer.
- 0:51SPEAKER_00: Most small businesses are, by Meta's definition, the preview of the terms counts, running ads, sending marketing messages, selling or listing products, managing a page, or using tools like Ads Manager or Meta Business Suite.
- 1:06SPEAKER_00: You're in.
- 1:07SPEAKER_01: So basically, every shop with a Facebook page.
- 1:11SPEAKER_00: Pretty much.
- 1:12SPEAKER_00: And here's the new noun we need a crawler.
- 1:15SPEAKER_00: A crawler is a program that visits websites on its own and copies what's on the pages.
- 1:20SPEAKER_00: Google uses one to build search results.
- 1:23SPEAKER_00: AI companies use them to collect text and pictures.
- 1:26SPEAKER_01: Got it.
- 1:27SPEAKER_01: A robot that reads your website so a person doesn't have to.
- 1:30SPEAKER_00: Right.
- 1:31SPEAKER_00: And the new terms say you authorize Meta or anyone acting from Meta to use crawlers on websites associated with your products or services.
- 1:40SPEAKER_00: They can cache what they find, which just means keep a copy and periodically pull updates.
- 1:46SPEAKER_00: Alright.
- 1:47SPEAKER_01: So why did this blow up?
- 1:48SPEAKER_01: I saw people losing their minds about it.
- 1:51SPEAKER_00: Because of one sentence.
- 1:52SPEAKER_00: The first version of the term said Meta could use content from those websites to provide, develop, improve, research, train, and maintain its products.
- 2:01SPEAKER_00: That word train is the AI word.
- 2:04SPEAKER_00: Training is how an AI model learns.
- 2:06SPEAKER_00: By reading huge amounts of material.
- 2:09SPEAKER_01: So the fear was Meta could read my website and feed it to its AI.
- 2:14SPEAKER_00: That was the fear.
- 2:15SPEAKER_00: A news site called Neiman Lab reported it on October 15th, and legal experts said the wording could let Meta scrape publishers' sites for AI training.
- 2:24SPEAKER_00: Publisher groups called it outrageous.
- 2:26SPEAKER_00: And then what did Meta do?
- 2:29SPEAKER_00: Meta said the crawling terms only ever applied to its ads and business products, and that it doesn't change how Meta uses web crawlers.
- 2:37SPEAKER_00: Then the sentence with the word train came out of the US English version.
- 2:42SPEAKER_00: The new wording says the crawling is to provide, develop, improve, research, and maintain advertising and business products.
- 2:50SPEAKER_00: So problem solved?
- 2:52SPEAKER_00: Not completely, and this is where I want to be careful.
- 2:55SPEAKER_00: Lawyers quoted by Neiman Labs said the word develop is still broad enough to cover training.
- 3:00SPEAKER_00: The new text doesn't say training is ruled out, and reporters who compared copies found the UK English version still had the word train, and one trade site said it still saw it on Meta's preview on October 18th.
- 3:12SPEAKER_01: So the wording has moved at least once in a week, and people don't fully agree on what's posted.
- 3:18SPEAKER_00: That's the honest summary, which is why the takeaway can't be trust the headline.
- 3:23SPEAKER_00: It has to be something you control.
- 3:26SPEAKER_01: Before we get there, what's Meta's own explanation for why they want to crawl my site at all?
- 3:32SPEAKER_00: A meta spokesperson said it's about pulling accurate information about products for sale on outside websites, so ads and recommendations on Facebook and Instagram show the right thing.
- 3:42SPEAKER_00: Think price, picture, whether it's in stock.
- 3:45SPEAKER_00: Which for a store selling stuff online honestly might help them.
- 3:50SPEAKER_00: It might.
- 3:50SPEAKER_00: If you sell products and run catalog ads, meta reading your product pages correctly is kind of the point.
- 3:57SPEAKER_01: Let me make this real for a service business.
- 4:00SPEAKER_01: Say I'm a pest control company.
- 4:03SPEAKER_01: Facebook page, I boost a post in the spring, my website has service pages, a few photos, a reviews page, and maybe a PDF price sheet somebody uploaded three years ago that I forgot about.
- 4:14SPEAKER_00: That forgotten PDF is the thing to think about.
- 4:17SPEAKER_00: If a page or file is public on your website, a crawler can usually reach it.
- 4:22SPEAKER_00: That's true for Google, it's true for AI companies, and now it's spelled out in your contract with Meta.
- 4:29SPEAKER_00: And the service pages and photos.
- 4:32SPEAKER_00: Those are marketing.
- 4:33SPEAKER_00: You want them seen.
- 4:35SPEAKER_00: For most small businesses, the website is a brochure, and somebody reading the brochure isn't a disaster.
- 4:41SPEAKER_01: Okay, so is this actually brand new or did they already have this?
- 4:45SPEAKER_00: Partly not new.
- 4:46SPEAKER_00: The self-serve ad terms in force now, from August 2025, already say that when you place an ad order, Meta can crawl the content on your domains for uses that include research and training.
- 4:58SPEAKER_00: So if you've been running ads, a version of this was already in there.
- 5:02SPEAKER_01: Huh.
- 5:03SPEAKER_01: So the scary sentence everybody found was kind of already in the ad terms.
- 5:09SPEAKER_00: For advertisers, yes.
- 5:11SPEAKER_00: What's new is that it's pulled into the main business terms, so it reaches anyone using Meta for business, not just people placing ad orders.
- 5:20SPEAKER_01: Can I opt out?
- 5:21SPEAKER_01: Is there a button?
- 5:23SPEAKER_00: No button.
- 5:24SPEAKER_00: The terms say if you keep using Meta's products for business after being notified, you've agreed.
- 5:29SPEAKER_00: The stated alternative is to stop all business use of Meta's products.
- 5:33SPEAKER_01: So the choice is accept it or shut down my Facebook page.
- 5:37SPEAKER_00: That's the choice in the contract.
- 5:39SPEAKER_00: And for most shops, deleting the page would cost far more than the crawler ever could.
- 5:44SPEAKER_00: So don't do that in a panic.
- 5:46SPEAKER_00: Alright, so what can I actually do?
- 5:49SPEAKER_00: Two things.
- 5:50SPEAKER_00: First, the website check.
- 5:52SPEAKER_00: Spend 15 minutes looking at your own site like a stranger would.
- 5:56SPEAKER_00: On Google, type site, a colon, your web address, and then the word PDF, and see what files come up.
- 6:03SPEAKER_00: Anything you wouldn't want copied, like an internal price sheet, an employee handbook, or a customer list, take it down or put it behind a real login.
- 6:12SPEAKER_01: And that's good to do no matter what Meta does.
- 6:16SPEAKER_00: Exactly.
- 6:17SPEAKER_00: Every crawler on the internet benefits from your forgotten files, not just Meta's.
- 6:23SPEAKER_00: And the second thing?
- 6:25SPEAKER_00: A file called robots.txt.
- 6:28SPEAKER_00: It sits on your website and tells crawlers which pages they're allowed to visit.
- 6:32SPEAKER_00: MetaZone crawler page says to block its crawlers, you add a disallow line for that crawler to robots.txt.
- 6:40SPEAKER_00: I don't know how to do that.
- 6:42SPEAKER_00: I'm guessing a lot of owners don't.
- 6:45SPEAKER_00: You don't need to.
- 6:46SPEAKER_00: Send your webperson one sentence.
- 6:48SPEAKER_00: Please check our robots.txt and tell me whether we're blocking Meta External Agent.
- 6:53SPEAKER_00: That's the Metacrawler their docs say is used for things like training AI models.
- 6:58SPEAKER_01: And if I block it, does that mess up my Facebook link previews, the little picture when I share my website?
- 7:05SPEAKER_01: Good question.
- 7:06SPEAKER_00: Link previews come from a different Metacrawler called Facebook External Hit.
- 7:10SPEAKER_00: So blocking the AI crawler alone shouldn't break your previews.
- 7:13SPEAKER_00: Don't block every metacrawler, or your shared links can look broken.
- 7:17SPEAKER_01: Is robots.txt a lock though?
- 7:20SPEAKER_01: Like does it actually stop them?
- 7:22SPEAKER_01: It's a request, not a lock.
- 7:24SPEAKER_00: Meta says content owners can use it to signal preferences, but the new terms don't promise to honor it, and a law professor quoted by Neiman Lab pointed out that following it is voluntary.
- 7:35SPEAKER_00: Meta's own docs also say two of its crawlers may skip it in certain cases.
- 7:40SPEAKER_01: So the real lock is not putting private stuff on a public website.
- 7:46SPEAKER_00: That's it.
- 7:47SPEAKER_00: Robots.txt is the polite sign on the door.
- 7:51SPEAKER_00: A real login is the lock.
- 7:54SPEAKER_01: What would you tell me to ignore?
- 7:57SPEAKER_00: Ignore the idea that Meta is breaking into your email or your bookkeeping.
- 8:01SPEAKER_00: These terms are about websites tied to your business.
- 8:04SPEAKER_00: And ignore the urge to quit Facebook over it.
- 8:06SPEAKER_00: The people with the most at stake here are publishers, and anyone whose website content is the actual product they sell.
- 8:13SPEAKER_00: That's the profit.
- 8:15SPEAKER_01: You didn't have to read Meta's contract or follow a week of back and forth.
- 8:20SPEAKER_01: You just need a 15-minute look at your own website and one question for your web person before October 30th.
- 8:29SPEAKER_00: One sentence to repeat.