{"id":18553,"date":"2026-08-26T18:43:03","date_gmt":"2026-08-27T01:43:03","guid":{"rendered":"https:\/\/jasonsblog.ddns.net\/?p=18553"},"modified":"2026-08-26T18:43:03","modified_gmt":"2026-08-27T01:43:03","slug":"the-meta-settlement-what-it-means-for-your-speech-and-privacy","status":"publish","type":"post","link":"https:\/\/jasonsblog.ddns.net\/index.php\/2026\/08\/26\/the-meta-settlement-what-it-means-for-your-speech-and-privacy\/","title":{"rendered":"The Meta Settlement: What It Means for Your Speech and Privacy"},"content":{"rendered":"\n

This is diabolical, and one angle on how they’re bringing about digital ID in the US. Just like Flock ALPR cameras with other technologies and companies blanketing the nation to track your movements, you’re going to be tracked online. And the tracking and censorship should let you know that Meta’s platforms are nowhere safe. And Meta has roots back to In-Q-Tel through Peter Thiel as I seem to remember, and the CIA put their own people within the management of Facebook, so is this manipulation of legal proceedings and subsequent agreement any surprise? Facebook was always a honeypot for you to provide all the information for the government to surveil you and your personal connections…<\/p>\n\n\n\n

https:\/\/reclaimthenet.org\/the-meta-settlement-what-it-means-for-speech-and-privacy<\/a><\/p>\n\n\n

<\/div><\/div><\/div>\n\n\n
Under the teen safety provisions is a court-supervised identity system for every Facebook and Instagram user.<\/h5>\n\n\n\n
\"\"\/<\/figure>\n\n\n\n

By Dan Frieth and Cindy Harper<\/p>\n\n\n\n

Today Meta agreed to make changes to Instagram and Facebook after many US states blamed it for creating dangerous products that harm children.<\/p>\n\n\n\n

The settlement brings an end to a major trial<\/a> that was taking place in California and will be a payout of up to $17bn from the big tech giant.<\/p>\n\n\n\n

The settlement is being heralded as a success in forcing the platform to bend to state demands on “safety” and introducing daily usage limits for teens and restrictions for nighttime use of social media.<\/p>\n\n\n\n

The states’ argument was that Meta purposefully designed products that were addictive, causing anxiety, depression, and suicide among teens.<\/p>\n\n\n\n

But read beyond the headlines and the feel-good announcements from the state attorneys general, and the 130-page proposed consent judgment is a very different story that has some big implications for free speech and privacy online.<\/p>\n\n\n\n

We obtained a copy of it for you<\/strong> here<\/strong><\/a>.<\/strong><\/p>\n\n\n\n

What it reveals is that the social media giant will have to introduce age assurance (that’s age verification, in plain English) for every user of Facebook and Instagram located in the states that are party to the settlement.<\/p>\n\n\n\n

And that’s not all. The deal also requires Meta to wire itself into the age signals that Apple and Google provide at the device level<\/a>, to get better at monitoring users to detect and merge accounts it believes belong to the same person, and to make sure that other major platforms \u2013 specifically Snap, TikTok, and YouTube \u2013 adopt the same practices.<\/p>\n\n\n\n

At first glance, it looks like it is about protecting children \u2013 time limits for teens, overnight blocks on the apps, no plastic-surgery filters. But the agreement goes far beyond that, and the age verification at the heart of it will touch every user. It also contains a wild clause in which both sides waive any right to argue that any of it is unconstitutional.<\/p>\n\n\n\n

Here are ten things you need to know<\/h2>\n\n\n\n
\"Silhouetted<\/figure>\n\n\n\n

ONE: Every user gets age-checked, not just teens<\/strong><\/p>\n\n\n\n

The age checks can’t be limited to teenagers \u2013 they apply to every user. That’s according to Section II.A.1 of the judgment, which requires Meta to apply its age assurance to every single social media user in the settling states.<\/p>\n\n\n\n

Section II.A.6.a.i further clarifies that this applies to both new and existing accounts, meaning every Facebook and Instagram user in those states.<\/p>\n\n\n\n

The methods are specified in Section II.A.3.b as ID verification and facial age estimation. If neither meets the required accuracy thresholds, Meta and the states will “meet and confer” to come up with an alternative \u2013 meaning that if the age checks fail, they are not abandoned; they are escalated.<\/p>\n\n\n\n

Two states, Illinois and Texas, have previously won record biometric-privacy settlements from Meta \u2013 $650 million and $1.4 billion, respectively. Illinois has now signed a judgment that names facial age estimation as a method to be used on every user in those states, the same kind of technology central to those biometric cases.<\/p>\n\n\n\n

The release at IV.A covers only the conduct alleged in the case, and does not extend to future scanning, so Illinois has not signed away its citizens’ biometric-privacy rights that it so often says it supports. But it has agreed to a deal that will see every one of those citizens age-checked by the very method that once cost Meta so much.<\/p>\n\n\n\n

TWO: It routes around the First Amendment, and the drafting proves it was deliberate<\/strong><\/p>\n\n\n\n

The age-verification consent judgement routes around the First Amendment.<\/p>\n\n\n\n

This has been achieved by including two paragraphs next to each other in Section X of the agreement, which is also repeated as IX.G of the consent judgment.<\/p>\n\n\n\n

X.E preserves Meta’s right to challenge any federal, state or local law, rule, regulation or guidance, whether presently enacted or enacted in the future.<\/p>\n\n\n\n

X.F, in the very next paragraph, has both parties waive any right to challenge or contest the validity of the agreement on any ground, including that any term is unconstitutional or preempted by or in conflict with any current or future law.<\/p>\n\n\n\n

These two paragraphs mean that Meta retains the right to sue over every age-verification law ever written, while giving up the right to challenge the one it just agreed to.<\/p>\n\n\n\n

This is particularly egregious given the record of similar legislation being blocked or permanently enjoined in courts, mostly on First Amendment grounds. Those laws were in Arkansas, Utah, Ohio, Texas, and Louisiana, while Virginia’s one-hour limit for under-16s was blocked in February<\/a>.<\/p>\n\n\n\n

It’s the same policy, but this time negotiated rather than tested in court, means there is no plaintiff and therefore no judicial scrutiny.<\/p>\n\n\n\n

THREE: Refuse to verify and you are treated as a child<\/strong><\/p>\n\n\n\n

One of the many ways Big Tech has been conditioning internet users to accept, and even welcome, digital ID, is by making the alternative very unappealing. On Facebook and Instagram, that means being treated as a minor.<\/p>\n\n\n\n

Section II.A.10 of the settlement spells it out: fourteen days after creating an account, a user who has not been assessed by an age assurance method is treated as a Teen User regardless of stated age. That means that they are subject to a number of restrictions: their account is capped at two hours of usage per day, they are cut off from midnight to 6am, they cannot see how many likes and other reactions their posts have received, and their feed is filtered.<\/p>\n\n\n\n

The only way to avoid this is to hand over ID, but nobody is forcing you to do that; you are simply downgraded to a minor’s version of the internet until you do. This is often referred to as “consent by attrition.”<\/p>\n\n\n\n

Those hoping that the settlement at least brings in some clarity and finality to this process will be disappointed, because, as usual, the devil is in the details. The judgment at II.B blocks non-parties from enforcing the settlement \u2013 and at IV.C.1(f) it preserves the separate claims private individuals may have.<\/p>\n\n\n\n

So, you cannot enforce this against Meta. The only route is Meta’s own appeals process, as stated at II.A.9 \u2013 and that only applies to adults who are wrongly identified as minors.<\/p>\n\n\n\n

FOUR: Meta has wanted this for years, and the contract now pays the states to force it on everyone else<\/strong><\/p>\n\n\n\n

Meta has for years been trying to get this exact policy of forcing age verification on everyone in the US, and now it has a contract that pays the settling states to make sure that happens.<\/p>\n\n\n\n

The giant has been very active in trying to get to this point. It lobbied both in Washington and in Brussels for rules that would force Apple and Google to accept age verification in their app stores<\/a>, and sued several US states to prevent them from passing their own age verification laws.<\/p>\n\n\n\n

But it was never that Meta was opposed to age verification per se \u2013 it was just opposed to carrying the liability for implementing it. And now, it has managed to get others to agree to take on that role, while it profits to the tune of billions of dollars.<\/p>\n\n\n\n

The contract, made public as part of the settlement with the state attorneys general, commits Meta to incorporating age signals from Apple and Google operating systems and app stores, with an Independent Auditor advising on the reliability of those signals.<\/p>\n\n\n\n

This is the same architecture that app-store statutes in Texas and California have. The Texas law, SB 2420, is already in force after the Fifth Circuit lifted an injunction against it in May<\/a>. The California law, AB 1043, will start requiring device-level age signals from 2027<\/a>.<\/p>\n\n\n\n

Meta’s contract came with an open letter<\/a> from attorneys general to TikTok and YouTube, urging them to adopt the same measures. The contract, meanwhile, defines “Core Industry Members” as Snap, TikTok, and YouTube in Section I.W.<\/p>\n\n\n\n

Three clauses in the settlement make this happen. Exhibit B guarantees Meta $1.166 billion a year, and makes another $502 million annual payment contingent on Snap, TikTok and YouTube joining the agreement as well, for a total of $16.68 billion over ten years, of which $11.66 billion is guaranteed and $5.02 billion is contingent on the other three companies signing up.<\/p>\n\n\n\n

The second clause, VI.D.3, states that if a state fails to hit its Contingent Monetary Payment Trigger during the term, the contingency payments are permanently forfeited and kept by Meta; but on a per-state basis, rather than the money being pooled. This means that every attorney general has a number riding on getting Snap, TikTok and YouTube to join the agreement \u2013 and will lose that money if they fail. For Virginia, this is $15.2 million per year.<\/p>\n\n\n\n

And the third, V.D, the Meta Injunctive most-favored-nation clause, says that if a settling state later cuts a deal with any Core Industry Member on terms better than Meta’s obligations under II.A, II.B and III, the state will consent to modify Meta’s agreement to match \u2013 with only the under-13 framework carved out. Paired with the Industry-Wide Adoption clause \u2013 under which rivals may exclude no features beyond messaging and long-form content without Meta’s written waiver \u2013 Meta holds both a floor and a ceiling on what its competitors can negotiate.<\/p>\n\n\n\n

And the “tell” is in the New SMP Entrant clause that automatically captures future platforms but excludes AI features, chatbots, and interactions with artificial intelligence. This is especially interesting given that Meta has spent this year in the press trying to position itself as an AI-first company.<\/p>\n\n\n\n

FIVE: A court is ordering Meta to link your accounts together<\/strong><\/p>\n\n\n\n

Meta is being forced by a court to get better at linking together the Facebook and Instagram accounts you may have opened under different identities.<\/p>\n\n\n\n

Section II.B.6.a commits Meta to using and improving its “soft-matching” models \u2013 including Single User Multiple Accounts, or SUMA, which draws on device IDs, phone numbers, and email addresses. The goal is to become industry-leading at identifying unlinked accounts that belong to the same user. The auditor will review the effectiveness of the models annually and recommend improvements.<\/p>\n\n\n\n

Alongside these soft matches, Meta also hard-links accounts that share the same login email or phone number.<\/p>\n\n\n\n

Section II.A.7 of the document caps age assurance attempts at three a day, four a month, and six over two years, counted collectively across all of a user’s hard-linked and soft-matched accounts. The section makes it clear that this applies to the matching model, that is, to accounts that are not linked by a user, but identified as belonging to the same user by Meta’s models.<\/p>\n\n\n\n

The same section also reveals that pseudonymity, the use of accounts not associated with a person’s real name, is now considered a compliance defect that the auditor is tasked with discovering.<\/p>\n\n\n\n

SIX: Being reported can put the burden of proof on you, and deletion spreads through who you know<\/strong><\/p>\n\n\n\n

And while we’re on the subject of reporting, it’s always good to remember that being reported can put the burden of proof on you, and even if you are not removed, the very process can be a form of punishment.<\/p>\n\n\n\n

The newly announced rules make things worse in this regard. Section II.A.6.b.i.C requires that both a simplified in-app reporting flow is implemented for those who suspect that a user is under 13, and that Meta’s existing web-based underage reporting tool is maintained, which can be used by anyone – i.e., not just those who have a Meta account.<\/p>\n\n\n\n

But the real-world effect turns on Section II.A.6.b.i.B, which provides that, absent reliable indicia of age, Meta shall presume the user is under 13, unless there is reliable evidence the account was reported by a malicious actor.<\/p>\n\n\n\n

You can treat that as a route in rather than a guaranteed trigger; the enforcement process is Meta’s own and the agreement does not define it. But the harm is intact either way. An anonymous account has no age indicia (that is what anonymity means) so it has nothing to rebut the presumption with, and mass reporting is already the standard technique used against dissidents, journalists, and pseudonymous critics.<\/p>\n\n\n\n

Then there is II.A.6.b.i.D: when an account is removed through under-13 enforcement, Meta must review that user’s friend networks to identify more under-13s. Deletion propagates outward through your contacts. That means more surveillance.<\/p>\n\n\n\n

SEVEN: It freezes Meta’s rules on lawful speech into a court order with targets<\/strong><\/p>\n\n\n\n

Meta is now under a legal obligation to maintain and improve censorship of speech that is already lawful under the company’s own rules.<\/p>\n\n\n\n

The obligation is spelled out in Section II.E. Section II.E.1 opens with Meta representing what it already does, and Section II.E.2 then requires it to maintain those features at least as effectively and to keep improving them.<\/p>\n\n\n\n

The giant is to keep and improve “features” that are already in place and are meant to “limit or interrupt” the display of content that is (already) considered lawful. However, it is content that falls into a number of categories, such as “sensitive” topics like discussions of mental health, body image, and fitness.<\/p>\n\n\n\n

The document states that Meta will have to maintain these “features” at least as effectively as they are now, and to continuously improve them. An independent auditor will assess how much of this content users are exposed to and report on it every year.<\/p>\n\n\n\n

Meta will also have to work to reduce the prevalence of this content, the document further states. The company is not being forced to do something new; it is being forced to keep doing what it is already doing, but to do it better.<\/p>\n\n\n\n

Section II.E.5.b requires Meta to decide on reports of potentially harmful content within six hours in at least 90 percent of cases. Given the volume of reports the company receives, this is a clear nod to the use of automation to make these decisions, and that usually means over-removal of speech.<\/p>\n\n\n\n

The categories are defined at Section I.BBB as Sensitive Aggregate Content: nutritional discussion, personal weight and fitness accomplishments, muscular physiques, non-invasive cosmetic procedures, body parts associated with idealization, and the depiction or discussion of personal experience living with mental or emotional distress.<\/p>\n\n\n\n

This means that a teenager writing about their own anxiety and feelings is producing speech that Meta is now under obligation to limit or interrupt.<\/p>\n\n\n\n

EIGHT: Suicide-prevention and recovery accounts get classified with the harm they exist to counter<\/strong><\/p>\n\n\n\n

Section I.E defines an Age Inappropriate Account two ways. The first is if the account “regularly” posts content in one of four categories: adult nudity and sexual activity, restricted goods and services, and suicide, self-harm or eating disorders. But an account can also be considered Age Inappropriate if its name, profile photo, or bio “indicates” that it is inappropriate for people under the age of majority (that is, minors) under Meta’s policies covering those same categories.<\/p>\n\n\n\n

This means that the content of the account is not necessary to determine that it should be blocked from all minors in the US, a bio will do. And this is where the four categories come in: suicide and self-harm, along with eating disorders, are right there with adult nudity, and restricted goods and services. In other words, accounts that exist to help with, support, and provide resources for the prevention of suicide, and to help people recover from eating disorders are lumped in the same category as content that promotes or glorifies these issues.<\/p>\n\n\n\n

This is a big deal. Section II.E.1.f blocks those accounts from being followed by, seen by, or interacted with by any teenager; they cannot be recommended to them, and these accounts are blocked from interacting with them at all. And there is no appeal. At least, the agreement does not provide for one. It does, however, explicitly provide for an appeal of a different sort: at section II.A.9, adults who are wrongly identified as minors can appeal this decision. But there is nothing about appealing being classified as an account inappropriate for minors.<\/p>\n\n\n\n

NINE: Your searches, likes and old group memberships mark you as suspicious<\/strong><\/p>\n\n\n\n

Your searches, likes, and even membership of groups that were later removed for violations of Meta’s policies will be used to determine if you are a “Potentially Suspicious Account.”<\/p>\n\n\n\n

Section I.TT defines a Potentially Suspicious Account by its interactions with other accounts, its searches for or interactions with violating content, including likes and saves, and its membership of communities Meta later removed for policy violations.<\/p>\n\n\n\n

Section II.F.4.a then requires Meta to warn the teen when a message thread opens with one of these accounts. Parents also get a separate notice under II.F.4.b, though that one is limited to accounts Meta thinks are likely to attempt financial blackmail or inappropriate interactions with children.<\/p>\n\n\n\n

What you searched, liked, saved, or joined feeds a status that is disclosed to strangers.<\/p>\n\n\n\n

And even if a user has done nothing wrong, they can still be marked as Potentially Suspicious if they were a member of a group that was removed for violating Meta’s policies, even if the user joined the group before it was removed.<\/p>\n\n\n\n

This means that the user has done nothing wrong, but is still penalized for something that happened after they had already joined the group.<\/p>\n\n\n\n

Meta also makes sure that parents are kept in the loop about their child’s activities on the platform. Section II.G of the agreement states that the company will share with a parent who has supervision over their minor child’s account the names of the accounts that the child is following and the names of the accounts that are following them.<\/p>\n\n\n\n

Also, every time the child first communicates with an adult, the parent will be notified with a link to the adult’s profile.<\/p>\n\n\n\n

Meta will alert parents when their child repeatedly searches terms related to suicide, self-harm or eating disorders.<\/p>\n\n\n\n

Third parties are identified to strangers without their consent, and the alerts land hardest on the very teens for whom home is the danger, the ones quietly looking for help who are outed to a parent by an automated notification.<\/p>\n\n\n\n

TEN: Meta gets to edit the public version of its own audit, and the evidence is destroyed afterwards<\/strong><\/p>\n\n\n\n

Meta’s own privacy and safety report is now “independent,” but the company still gets to control what the public gets to know about it.<\/p>\n\n\n\n

Section III.J requires the auditor to publish an executive summary of each final report, but the same clause lets Meta and the State Committee review and comment on a draft of that summary, including to ensure the exclusion of anything nonpublic, proprietary or confidential.<\/p>\n\n\n\n

So the company under audit gets editorial input on the public account of its own compliance, and the underlying record does not survive. Section III.I.1 lets Meta designate materials confidential; Section III.I.2 requires everything the auditor obtained from Meta to be returned to Meta or destroyed when the auditor’s term ends; and Section III.G.7 requires any state that receives a court order or records request covering audit materials to give Meta at least ten days’ notice, so Meta can go to court first.<\/p>\n\n\n\n

You’re going to see a lot of coverage about this story over the next few days. Attorneys general will put out their press releases and call it a win for children. Meta gets to take out ads in newspapers and present itself as the company that finally did something.<\/p>\n\n\n\n

What will get ignored are the implications of these settlements. It means age checks on every adult, a court order requiring Meta to get better at surveilling users and linking your accounts together, deletion that spreads through your friend list, an audit summary the audited company helps edit, and contracting away the right to call any of it unconstitutional (which much of it likely would be if states tried to enforce it through law.)<\/p>\n\n\n\n

Users are on the brink of being identified, classified, and monitored on a scale no legislature would have survived voting for. Meta will come out ahead on all of it. It pays a decade of installments out of a single year’s profit, freezes its existing content rules into a judgment it helped draft, holds a most favored nation clause over whatever deals its competitors manage to cut, and watches attorneys general go after Snap, TikTok and YouTube.<\/p>\n\n\n\n

At the end of it all, the states get something better than money. They get the identity infrastructure that some courts have blocked them from legislating, delivered voluntarily by the company that was suing them over it, with the constitutional challenge signed away before anyone could bring it.<\/p>\n","protected":false},"excerpt":{"rendered":"

This is diabolical, and one angle on how they’re bringing about digital ID in the US. Just like Flock ALPR cameras with other technologies and companies blanketing the nation to track your movements, you’re going to be tracked online. And the tracking and censorship should let you know that Meta’s platforms are nowhere safe. And […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[6,7],"tags":[],"class_list":["post-18553","post","type-post","status-publish","format-standard","hentry","category-tech","category-world"],"blocksy_meta":{"styles_descriptor":{"styles":{"desktop":"","tablet":"","mobile":""},"google_fonts":[],"version":7}},"featured_image_src":null,"author_info":{"display_name":"Jason","author_link":"https:\/\/jasonsblog.ddns.net\/index.php\/author\/jturning\/"},"_links":{"self":[{"href":"https:\/\/jasonsblog.ddns.net\/index.php\/wp-json\/wp\/v2\/posts\/18553","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/jasonsblog.ddns.net\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/jasonsblog.ddns.net\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/jasonsblog.ddns.net\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/jasonsblog.ddns.net\/index.php\/wp-json\/wp\/v2\/comments?post=18553"}],"version-history":[{"count":1,"href":"https:\/\/jasonsblog.ddns.net\/index.php\/wp-json\/wp\/v2\/posts\/18553\/revisions"}],"predecessor-version":[{"id":18554,"href":"https:\/\/jasonsblog.ddns.net\/index.php\/wp-json\/wp\/v2\/posts\/18553\/revisions\/18554"}],"wp:attachment":[{"href":"https:\/\/jasonsblog.ddns.net\/index.php\/wp-json\/wp\/v2\/media?parent=18553"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/jasonsblog.ddns.net\/index.php\/wp-json\/wp\/v2\/categories?post=18553"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/jasonsblog.ddns.net\/index.php\/wp-json\/wp\/v2\/tags?post=18553"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}