Rhode Island sued Meta on Thursday, October 22, 2026, accusing the company of profiting from scam ads on Facebook, Instagram and WhatsApp. The complaint, filed in Providence Superior Court, goes further than earlier cases by alleging that Meta’s own generative AI ad tools helped scammers build their campaigns. Rhode Island is the first state to bring a case over this conduct, and the suit puts Meta’s AI ad products under state consumer-protection scrutiny.
Reuters reported the filing on October 23 (Reuters via Investing.com). Meta says it will fight the case and calls it “unfounded.” Here is what the complaint says, where the biggest numbers come from, and what happens next.
Rhode Island sues Meta over scam ads: what the complaint alleges
Attorney General Peter Neronha’s office accuses Meta of unlawfully profiting from scam advertising. According to FourWeekMBA’s summary of the case, the complaint says Meta failed to curb ads for fraudulent products and financial scams, despite publicly claiming sophisticated prevention efforts.
Neronha put the core claim in one sentence: “As alleged, Meta knowingly exposed vulnerable users to fraud and scams, and made enormous profits in doing so.” The words “as alleged” matter. These are accusations in a complaint, and no court has ruled on them.
The suit also says Meta sometimes charged suspected scammers extra instead of blocking them. That detail ties the enforcement problem directly to revenue, because a fee for continuing to run an ad turns uncertainty into income.
The complaint also offers concrete examples. One is a single ad with more than 2,000 versions active in the Meta Ad Library, promising to reverse Alzheimer’s in 45 days.
On remedies, the lawsuit seeks civil penalties for each alleged violation of state law, restitution for consumers, disgorgement of ill-gotten gains, and a court order requiring Meta to address scam ads rather than profit from them, according to Engadget’s report. Disgorgement is the most financially pointed request, because it targets the money itself rather than only the ads.
How Meta’s AI ad tools factor into the scam ad claims
The AI allegation is what sets this case apart. Earlier scam-ad fights mostly asked whether a platform removed bad ads fast enough. Rhode Island argues that Meta’s AI ad creation tools went beyond passive hosting and, in many cases, actively assisted scammers with their fraudulent marketing, according to the Northwest Arkansas Democrat-Gazette.
Earlier reporting adds context. A 2025 Reuters investigation found that Meta used AI to generate and test deceptive ad variations and steer them toward users who had clicked on scam ads. The Santa Clara County complaint drew on that investigation, as Fortune’s coverage of the county suit describes.
The legal question is whether a court sees a difference between a platform hosting an advertiser’s content and a platform whose own tools help write, test or target it. If the judge sees that gap as meaningful, the case could reach well beyond scam ads, into any ad product where the company’s software shapes the creative. If the judge does not, the AI angle could fade. Either way, Rhode Island has to prove what its complaint alleges.
The same question is spreading across the industry. OpenAI has widened its own advertising business, with ChatGPT ads now running in 89 countries. Every company that sells AI-assisted ad creation will be watching how this complaint treats the line between a tool and an advertiser.
The scam ad numbers in dispute
Several figures in the coverage are large, and most come from contested sources. The table below separates what each number measures from how solid its basis is.
| Figure | What it measures | Where it comes from | Status |
|---|---|---|---|
| About 15 billion “higher risk” scam ads per day | Ads displayed to users daily | Internal Meta documents, as reported by Reuters in 2025 | Contested; Meta disputes the reporting |
| About 10% of 2024 revenue | Revenue tied to fraudulent e-commerce, dubious health products and other banned marketing | Same internal documents | Contested |
| Around $16 billion | Projected 2024 revenue from scam and banned-goods ads | TheNextWeb, attributing the figure to a Meta projection | Single-source attribution |
| About $7 billion a year | “Violating revenue” cited in the Santa Clara County complaint | Santa Clara County suit | Allegation |
| More than 159 million scam ads removed | Enforcement volume | Meta, as reported by Engadget | Company claim |
Treat these as allegations and company statements, not findings. The 15 billion and 10 percent figures trace back to internal documents that Reuters reported on in 2025, and Meta has disputed that reporting. The $16 billion figure appears in one outlet’s report.
Scale matters, too. Meta made more than $200 billion in revenue in 2025, according to Al Jazeera, so the financial stakes depend on penalties a court has not yet set.
How Meta responds and where its enforcement line sits
Meta says it will fight the suit and calls it “unfounded.” Spokesperson Andy Stone told Reuters that the attorney general “refused to engage with us” and “has instead hired an out-of-state private law firm to concoct a lawsuit,” as Engadget reported. Meta is framing the case as a dispute over process and motive, not over whether fraud reached its platforms.
The enforcement threshold is the most specific allegation. According to Meta’s internal documents, as cited in the Santa Clara County complaint, Meta banned marketers only when it was 95 percent certain they were committing fraud. Suspected scammers below that line were charged a premium fee to keep running their ads.
That policy turns a gray area into a revenue stream, and it is the detail most likely to shape how judges and regulators view the company’s motives. The brief does not include a direct Meta response to the threshold claim, and Meta has disputed the Reuters reporting the complaint relies on.
Legitimate advertisers have a stake too. The Santa Clara complaint argues that when fraudulent advertisers flood ad auctions, honest businesses pay more to reach the same users.
Why older adults and advertisers sit at the center
The complaint highlights consumers, especially older adults. It alleges that scam ads often targeted elderly people, citing an investigation that identified nearly 94,000 scam ads on Meta platforms, as NewsBytes reported.
Scam ads that promise cures or quick financial returns work by building trust and urgency. That makes the age profile of victims central to the consumer-harm argument, and it gives the state a clear group to point to in court.
How to report scam ads on Facebook and Instagram
Readers can act now, whatever happens in court.
- Treat ads for miracle cures, crypto offers and celebrity giveaways as suspicious until you verify them outside the ad.
- Use Meta’s reporting tools on the ad or post to flag it, even if you saw it only once.
- Ignore urgent money requests that show up in your feed. Confirm them through a phone number or contact you already have.
- If you already sent money to a scammer, contact your bank or card issuer right away.
What comes next for Rhode Island’s Meta lawsuit
No court schedule, hearing date or response filing appeared in the sources reviewed, so the next signals will come from court records. Watch for Meta’s formal response. It will show whether the company contests the AI-tool allegations directly or sticks to the process dispute it has raised publicly.
Watch, too, whether Rhode Island can back its AI claims with evidence, since the allegations come from a complaint and not a finding. And watch whether other attorneys general follow. Rhode Island is the first state, but the Consumer Federation of America and Santa Clara County have already sued, so pressure is building from several directions.
The next few months will show whether this case becomes a template for state enforcement against AI-assisted advertising, or narrows into a fight over one company’s enforcement policy.
