The trial that could quietly rewrite meta advertising policies

The Oakland trial marketing directors can't scroll past#

A fox does not need to catch every rabbit in the hedgerow to know the weather has turned. It reads the wind, adjusts its prowl through the thicket, and moves before the storm lands, well before the scent of change reaches the open ground. That is the posture every paid social advertising lead should be borrowing right now, because a federal courtroom in Oakland has spent this week hearing exactly how Meta built Facebook and Instagram, and the answer it settles on will not stay inside the courthouse.

Jury selection opened on 12 August 2026 in the courtroom of US District Judge Yvonne Gonzalez Rogers, and opening statements began on 18 August, with the presiding judge telling the pool of eight jurors the trial should run about six weeks Courthouse News, 2026. This is the first jury trial to reach a verdict inside a sprawling multidistrict litigation that now consolidates more than 3,100 lawsuits over youth social media harm, and the states involved are treating it as the bellwether case for everything that follows Tech Policy Press, 2026. It is also, functionally, the first time meta advertising policies on youth data will be tested in front of a jury rather than negotiated behind closed doors.

A coalition of 33 attorneys general first filed the federal case on 24 October 2023, accusing Meta of designing addictive features into Instagram and Facebook while publicly understating the harm those features caused, in violation of the Children's Online Privacy Protection Act, California's False Advertising Law and California's Unfair Competition Law California DOJ, 2023. Only four of the states in that coalition, California, Colorado, Kentucky and New Jersey, are actually trying the case in Oakland this month.

The other twenty-five are watching from the wings, either negotiating settlements of their own or waiting for a later trial date, which is exactly the kind of composition marketers should sit with before assuming meta advertising policies are someone else's problem. Bonta laid out the case in full the day before opening statements, describing a company that "designed Facebook and Instagram to keep kids on the platforms longer and longer" California DOJ, 17 August 2026.

The number that has made every marketing director's inbox lurch is the damages figure: up to $1.4 trillion, a sum Meta itself disclosed in a legal filing and one that sits close to the company's entire market value Tech Policy Press, 2026. Nobody credible expects a court to award anything near that ceiling; the figure exists because the states are pricing statutory penalties per violation across millions of accounts, and the multiplication runs wild long before a jury gets a say.

What matters for a paid social advertising budget is not the headline trillion, it is the direction of travel: courts keep siding with the states on the underlying facts, and every ruling narrows what Meta can plausibly claim about how its products are built.

Mark Zuckerberg and Instagram head Adam Mosseri are both expected to be called to testify Tech Policy Press, 2026, which turns this from a procedural filing into six weeks of sworn testimony about design decisions, internal research and what the company knew and when. Depositions like that have a habit of surfacing the sentence a compliance team spends the next year explaining, and the sentence in question here is very likely one buried inside meta advertising policies rather than in the platform's public relations copy.

What the states allege Meta built, and why it touches your targeting#

Strip the legal language back and the states' case rests on three claims, and every one of them brushes against the machinery brands rent when they buy reach on Meta. First, that specific design choices, infinite scroll chief among them, were built to keep young users producing engagement signals rather than to serve them well Courthouse News, 2026. Second, that Meta collected data on under-13 users without verifiable parental consent, a straight COPPA violation. Third, that the company's public statements about platform safety misrepresented what its own research showed, the deceptive-advertising thread that pulls in California's False Advertising Law and Unfair Competition Law.

That third claim is the one worth sitting with if you run paid campaigns on the platform, because meta advertising policies and Meta's public safety claims are not separate conversations, they are the same conversation wearing different clothes. A company defending its own marketing claims in federal court is, structurally, in the same position as an advertiser defending a claim on its landing page.

The Federal Trade Commission's endorsement and advertising guidance sets a plain bar for that kind of claim: it must be truthful, substantiated before it is made, and not likely to mislead a reasonable consumer Federal Trade Commission. If a court finds Meta fell short of that bar on its own safety messaging, the same ftc advertising guidelines apply to every brand whose ad copy leans on Meta-provided targeting or safety assurances it has not independently checked.

The June ruling that kept the case alive#

On 30 June 2026 the court fully denied Meta's motion for summary judgment, the procedural move that would have ended the case before trial, and agreed with California that Meta had not adequately secured parental consent under COPPA California DOJ, 2026. "California is proud to help lead this nationwide reckoning, and we look forward to holding Meta fully accountable at trial," Bonta said at the time, and that ruling is why Oakland is happening at all rather than settling quietly the way most corporate litigation does.

Meta's own defence, published on its safety pages, is worth reading in full rather than through a headline. The company calls the states' claims "unsubstantiated" and says the litigation relies on "cherry-picked quotes" that paint "an intentionally misleading picture," pointing instead to Teen Accounts with built-in restrictions that 97% of 13 to 15 year olds have stayed inside, content limited to a 13-plus movie rating by default, and more than $30 billion invested in safety teams and technology over the past decade Meta, 2026.

Whatever the jury makes of that record, it is the clearest public statement of what Meta considers a defensible standard for youth-facing product design, and it is worth benchmarking your own creative against the same standard rather than assuming Meta's review process will catch what a court might not.

Those figures come from two cases already decided inside the same multidistrict litigation. In March a Los Angeles jury found Meta 70% liable and Google's YouTube 30% liable for a young plaintiff's platform addiction, awarding $6 million in compensatory and punitive damages ABC7 Los Angeles, 2026.

Days before Oakland's jury selection began, a New Mexico court ordered Meta to pay a further $567 million into a treatment and prevention fund on top of an earlier $375 million penalty, a combined $942 million, and required Meta to make under-18 accounts private by default, pause push notifications to minors between 10pm and 7am, and cap their monthly usage near ninety hours PBS NewsHour, 2026. "Today's decision is a victory for every parent who has worried about what social media is doing to their child," New Mexico Attorney General Raúl Torrez said when the ruling landed.

Those are product mandates a court can impose regardless of what any single advertiser intended, and they are the clearest signal yet that meta advertising policies on youth data are heading toward tighter defaults, not looser ones.

That question, asked flatly by a Reddit user sharing the wire copy the day opening statements landed, is a fair proxy for how much of the general public actually understands about the law underpinning this case. Most of your customers will not know COPPA from a compliance acronym soup. They will, however, notice a headline that says Meta was fined again, and they will read your brand's Meta ads through that headline whether the connection is fair or not.

The research the jury will hear, and what it means for your creative#

Courtroom argument aside, the underlying science is not new to anyone who has tracked the last three years of adolescent mental health research, and it is worth knowing what it actually says rather than the flattened version that survives a headline, and rather than the flattened version meta advertising policies debates usually settle for. The largest relevant dataset is the Adolescent Brain Cognitive Development Study, a long-running US cohort that has followed thousands of children from around age ten into their teenage years.

What the ABCD data actually shows#

A JAMA study published on 18 June 2025 followed 4,285 adolescents in that cohort from age eleven to fifteen and sorted their social media habits into distinct trajectories rather than treating screen time as one number Xiao et al, JAMA, 2025. Just under a third of the group, 31.3%, showed an "increasing" addictive-use pattern, and a further 9.6% ran "high" from the start.

By the fourth year of follow-up, adolescents on the increasing trajectory carried a risk ratio of 2.14 for suicidal behaviours compared with low-use peers, and those on the high trajectory carried a risk ratio of 2.39. Suicidal ideation moved in the same direction, though less steeply, at risk ratios of 1.46 and 1.51.

Across the whole cohort, not just the highest-use group, 17.9% of the 4,285 adolescents reported suicidal ideation by that fourth-year check-in.

The detail that should reshape how you think about creative and cadence is the one the study's authors flagged themselves: baseline screen time on its own showed no meaningful association with any of those outcomes. It was not hours logged that predicted harm, it was the pattern of compulsive, hard-to-stop use, the kind that platform mechanics can either dampen or encourage.

That distinction is precisely what a narrative review published in Cureus in February 2026 goes on to unpack, naming "algorithmic personalisation, notifications, and reward-based feedback loops" as the specific design elements that increase engagement intensity and emotional arousal, and singling out "infinite scroll, streaks, and push notifications" as features engineered for exactly that effect Sharma & Sharma, Cureus, 2026. The same review puts adolescent depression prevalence at roughly 5 to 8% in early adolescence, climbing to 15 to 20% by late adolescence, a curve that tracks the years most teen-reachable ad accounts are actively targeting.

None of that research proves any single brand's campaign caused harm to any single teenager, and the review's own authors are careful to describe the individual-level effects as modest, gaining their weight from how many adolescents are exposed at population scale. But it is the evidence base the plaintiffs are building their case on, and it is a fair description of the terrain your creative sits inside once it reaches a teen or near-teen audience through Meta's placement stack.

The question worth asking your creative team is not whether your ad caused the underlying compulsive pattern, it obviously did not, but whether your format, frequency and targeting are actively feeding a loop the platform is already under a court order to slow down elsewhere. This is the research meta advertising policies will keep being measured against long after the Oakland jury goes home.

This is also where the earlier ruling on ranked, personalised feeds becomes relevant to Meta specifically. A separate US court held earlier this month that a minor's ranked, algorithmically personalised feed is conduct, not protected speech, which is the same design-not-content framing running through the Oakland trial, and it is a shift folkfox has already tracked for what it means for social media regulation and paid social reach. Read alongside the ABCD figures, the throughline is consistent: regulators and courts are increasingly treating the ranking and recommendation engine itself, not any individual piece of content, as the thing under review.

Auditing your meta advertising policies exposure now#

None of this means panic-pulling budget out of Meta's ad manager before the jury has even heard closing arguments. It means running the audit you would run before any platform faced a regulatory shift this size, treating your current meta advertising policies compliance as a working document rather than a settled fact, and running it now while the verdict is still weeks away rather than after a ruling forces your hand.

The audit below is short by design, five checks against your current meta advertising policies, not fifty. Start with who your creative can actually reach. Meta's own Teen Accounts default to restricted contact and content settings, and 97% of 13 to 15 year olds have stayed inside them Meta, 2026, which means most brands running broad-reach campaigns are already being filtered away from younger teens by the platform's own guardrails. Confirm that with your own account data rather than assuming it: pull your age-bracket delivery breakdown and check it against your intended audience, the same way folkfox has flagged brands checking their reporting after Meta's own breakdown fields went quiet quietly, without warning earlier this month.

Second, audit the claims in your ad copy against the same standard the FTC applies to any endorsement or influencer content: is every material claim about safety, results or product benefit something you can substantiate on request, disclosed clearly, and free of the kind of "cherry-picked" framing Meta itself now stands accused of using Federal Trade Commission? The ftc advertising guidelines were written for exactly this kind of scrutiny, and a compliance review that was optional eighteen months ago is closer to mandatory now that a regulator's own trading partner is defending near-identical claims in federal court.

That fourth row matters more than it looks. Meta updates its meta guidelines for ads on a rolling basis, and a policy team defending itself in federal court has every incentive to tighten its own advertiser-facing rules faster than usual over the coming months, both to demonstrate good faith to the jury and to get ahead of whatever remedy a verdict might impose. Treat the current policy text as a snapshot, not a settled reference, and archive it against a review date so you can prove when your account last matched the live rules.

Most brands do not have a standing owner for social media advertising compliance the way they have an owner for paid search or email. This trial is a reasonable prompt to name one, even if it is a shared responsibility split across the five rows above rather than a single new hire.

Third, look at what New Mexico's court already ordered Meta to build and ask whether your own creative assumes the old defaults. Private-by-default under-18 accounts, notification pauses overnight, and a monthly usage cap near ninety hours are not hypothetical product changes, they are a live legal remedy in one state already PBS NewsHour, 2026, and folkfox has tracked the same court's earlier delivery-window order and what it means for paid social reach specifically. A campaign built around late-evening delivery windows or frequency that assumes an always-on teenage audience is betting against the direction every recent ruling has pushed.

Diversifying paid social advertising without panicking#

Diversification is the sensible response to genuine platform risk, and it is also the response every media planner reaches for regardless of the evidence, so it is worth being precise about what this trial actually changes for your paid social advertising mix and what it does not.

It does not change Meta's reach, its auction dynamics or its measurement stack this quarter. Whatever the jury decides, any product remedy will take months to specify and longer to implement at platform scale, the way New Mexico's mandates carry a two-year runway for some provisions PBS NewsHour, 2026. What it does change is the confidence you should place in Meta's current policy language holding steady, and the cost of being caught flat-footed if a remedy lands mid-quarter. That is a budgeting question, not a channel-abandonment question.

Where a genuine channel decision belongs is in your test budget, not your core allocation. folkfox has already tracked how brands are weighing behavioural targeting on Meta against newer conversational surfaces and how platforms are quietly treating AI-made creative as a distribution problem in its own right, and the same discipline applies here.

Build the second channel's creative and measurement now, while it is a controlled experiment, rather than for the first time under pressure the week a verdict lands. The brands that handled Pinterest's European slowdown well were the ones that already had a tested alternative sitting on the shelf, not the ones improvising one after the fact.

California's Department of Justice is not slowing down elsewhere either. The same office pursuing Meta has also been moving fast on other consumer-protection fronts this year, from data practices to ticket-resale enforcement under AB 1349, which is a useful reminder that this AG's office treats platform-scale consumer harm as a standing priority, not a one-off case. Meta advertising policies are simply this month's example of a pattern worth watching generally.

The point of this piece is not that meta advertising policies are collapsing, they are not, and Meta will keep selling reach through and past this verdict regardless of the outcome. The point is that meta youth safety has stopped being a brand-safety footnote and become a live legal variable, one with a court date, a named judge and sworn testimony from the platform's own leadership.

A media plan that treats it that way, with a documented audit, a tested second channel and a monthly policy recheck, will outlast whatever the jury decides. A plan that only notices when the verdict lands will be auditing under deadline instead of on schedule, and that is always the more expensive way to outfox a moving target.

If you want that audit built rather than described, that is the discipline behind folkfox's paid social advertising services, applied with the same rigour we bring to brand strategy work across regulated and reputation-sensitive accounts.

Frequently asked questions#