In addition to paying out up to $18 billion and adding child safety measures, Meta’s settlement agreement with attorneys general from 29 states includes an interesting provision: the states have agreed not to sue Meta under existing child safety laws over its retention and use of children’s data.

That permission is being granted for the limited purpose of training and testing Meta’s age-assurance model and includes guardrails, but it’s a curious policy decision to make in a case centered on child safety, and one that could be difficult to properly enforce.

As specified in the settlement agreement, Meta must develop, train, and begin testing a model designed to detect which users on Meta’s platforms are under the age of 13. This must be done within a year of the document’s effective date. (While the agreement doesn’t specify that the model has to be AI-based, Meta’s current age-detection tools are powered by AI technology.)

Under U.S. child safety law, COPPA (the Children’s Online Privacy Protection Act), typically requires that websites and apps limit the collection and retention of children’s personal information. Meta’s settlement agreement says that Meta shouldn’t need to violate COPPA to train or implement its age-assurance models. However, the agreement also says that the state AGs have agreed “fully, finally, and forever” not to bring any past, present or future COPPA claims — or claims under similar state laws — related to Meta’s use of children’s data.

The agreement makes clear that Meta can’t use data from users under age 13 for ad targeting, marketing, or algorithmic optimization.

Meta’s request for legal protection, and the state AGs’ willingness to grant it, isn’t unreasonable, says Philip N. Yannella, a partner at law firm Blank Rome and co-chair of its Privacy, Security & Data Protection practice. “These kinds of data minimization guardrails are pretty typical for privacy compliance: e.g., verifying compliance with deletion requests,” he said, though he noted a caveat: COPPA is a federal law primarily enforced by the FTC, not the states, so it’s unclear whether the FTC, which isn’t a party to this settlement, has separately agreed to the same compromise.

It can be difficult for companies to keep data technically and organizationally isolated from the rest of their systems. Yet Meta is being asked to do just that — to isolate its understanding of children’s behavior signals and other data and use it solely for detecting and removing under-13 users. Fortunately, an independent auditor will be involved in monitoring Meta’s compliance with the settlement so we don’t only have to rely on Meta’s word.

Policing this limitation could be complicated. The data could hypothetically feed into other Meta systems over time, or could raise questions over whether the data, signals, or insights derived from it are being used elsewhere within the company. What’s not clear from the agreement is what data Meta will retain for training the model, how much behavioral information that may include, or how long it will retain the data. We also don’t know how these models will change in the future as Meta meets the settlement’s terms.

Barring state AGs from raising COPPA or similar state-law claims over this use of children’s data in the future could complicate the legal avenues states can pursue if questions arise around how Meta is using the data.

That doesn’t prevent them from pursuing legal claims, notes Joshua Wurtzel, a partner at Schlam Stone & Dolan LLP. “If Meta uses the data outside those lines, the release and covenant not to sue don’t apply,” he said. But those legal disputes could still be complicated, since they’d hinge on whether Meta’s use of the data fell within the settlement’s terms.

Peter Jackson, a Data & IP attorney at Greenberg Glusker LLP, agrees, saying the carve-out here could “disincentivize future enforcement actions.”

“The Settlement Agreement’s age-assurance measures bear all the hallmarks of a heavy, and perhaps hasty, negotiation,” he says.

The decision also touches on a broader question that’s been coming up across the AI industry lately, especially as more AI agents are being developed to help consumers with various tasks. The systems often require significant access to users’ personal data to work well. Similarly, Meta may need deep insight into children’s use of social media use in order to identify which accounts belong to young people.

The game of musical chairs for AI executives continues. Barret Zoph, a co-founder of the AI startup Thinking Machines who left the company earlier this year to rejoin OpenAI, has found yet another job.

Zoph spent two years at OpenAI and left in October 2024 to co-found Thinking Machines with Mira Murati, who had left the AI lab the month prior. In January, Zoph and another Thinking Machines co-founder, Luke Metz, quite dramatically departed from the startup to return to OpenAI.

His return didn’t quite stick. Zoph spent only five months at OpenAI, where he was tasked with heading AI enterprise sales. He left the company in June. And now we know where he landed.

Zoph has taken a position as vice president of research at Google (which happens to be another company where he previously worked). “We look forward to Barret returning to Google and bringing his RL and post-training expertise to Gemini,” a Google spokesperson told the Wall Street Journal.

TechCrunch reached out to OpenAI and Google for more information.

It’s not always easy to divine why tech executives seem to be spending less time in their roles. The turnover rate in the AI industry is high, and it’s been especially high at OpenAI — a company that, despite readying itself for an IPO and being one of the most powerful presences in the tech world, has lost a lot of critical staff over the last eight months. High-level executives — from the departure of its COO to the recent loss of one of its top data center execs — has left onlookers scratching their heads.

YouTube announced on Thursday that eligible creators in the U.S. can now tag Amazon products in their content and receive a cut of sales. Creators can link Amazon products in their shorts, long-form videos, and livestreams.

The update turns product recommendations into a more direct revenue stream for creators, and for Amazon, the move puts its massive online marketplace inside one of the most popular video platforms.

Although YouTube already runs a Shopping affiliate program with participating retailers, the addition of Amazon’s vast catalog essentially allows creators to recommend a variety of products through a native integration. By bringing Amazon into the Shopping Affiliate Program, creators no longer need to paste Associates links in the video’s description, then hope viewers copy the link when they make a purchase.

Instead, YouTube says Amazon will provide it with a curated catalog of highly requested and trending products that creators can tag in their videos. If a creator can’t find a specific product that they want to tag, they can request to add it by reaching out to YouTube Support.

The feature also includes auto-tagging support. If enabled, YouTube’s systems can automatically review a creator’s recent uploads to identify and tag eligible Amazon products.

YouTube notes that creators won’t see breakdowns for specific products or individual videos in their analytics, but that they’ll see their overall daily earnings in YouTube Studio. If a viewer ends up returning an item, that commission will be deducted from the creator’s balance.

While only eligible U.S. creators can currently tag Amazon products, those tags can be seen globally. YouTube says it may automatically match a tagged product with a trusted local merchant offer to allow creators to earn commissions on eligible international purchases. If a local merchant isn’t available, the tag will route viewers to the Amazon U.S. website, where commission will be earned if a purchase is finalized on the U.S. site.

To be eligible, creators must be enrolled in YouTube’s Partner and Shopping Affiliate programs, have an active Amazon Influencer or Associates account, and link it to their YouTube channel.