Rhode Island Is First State to Sue Meta Over Scam Ads

Rhode Island Attorney General Peter F. Neronha sued Meta on 22 October 2026, alleging the company failed to curtail advertisements for fraudulent products and financial scams on its platforms “despite representing to the public that it engages in sophisticated efforts to prevent such ads,” according to his office’s release. The release says “Rhode Island is the first state to file a lawsuit against Meta for this alleged behavior, following a similar lawsuit brought by Santa Clara County in California.”

The complaint, filed in Providence County Superior Court under the state’s Deceptive Trade Practices Act, alleges that Meta “knowingly helped create and optimize these ads.” Meta spokesperson Andy Stone told Reuters that the company has removed more than 159 million scam ads from its platforms, Engadget reported, and said “The attorney general refused to engage with us on ways we protect Rhode Islanders from scams.”

What the Complaint Alleges

• The money. Citing Reuters, the complaint says Meta projected that ads for scams or restricted goods would account for about 10% of its 2024 revenue, or $16 billion, and that ads that “present higher legal risk” alone yielded $7 billion in annual revenue.
• The tools. The complaint describes Meta’s dynamic creative, Advantage+ and a text-generation feature it calls “a generative AI feature”, and says Meta uses “highly engineered generative AI models” to identify the users most likely to be moved to action by any particular ad, including scam ads. The release says Meta provided “design assistance, audience-selection tools, targeting capabilities, optimization systems” and related services to such advertisers.
• The surcharge. According to reporting cited in the complaint, at lower levels of confidence that an ad was fraudulent, “Meta merely charged a premium for the ad instead of removing it.”
• Verification. The complaint says Meta’s advertiser verification program is voluntary and fee-based, and that, according to internal documents reviewed by Reuters, Meta estimated verification would cost $2 billion, “just over 1% of Meta’s advertising revenue.”
• Rhode Island examples. The complaint says there were more than 2,000 versions of one ad active in the Meta Ad Library promising to reverse Alzheimer’s in 45 days, and that some advertisers named in a Tech Transparency Project report spent more on Meta ads over a several year period than large brands like Walmart or Pfizer.

Source: Rhode Island’s complaint against Meta, filed 22 October 2026 (public version). These are allegations and figures the complaint attributes to other reporting; a court has not ruled on them. The figures measure different things and are not added together.

What Was Posted on X

@choblin29’s post on X on 22 October said Rhode Island’s complaint cites 2,000-plus active Alzheimer’s-cure ads and claims some scam advertisers outspent Walmart or Pfizer on Meta. It also quotes what it calls an internal Meta review comparing Meta with Google.

What Meta Says

In its statement to Reuters, as quoted by Engadget, Meta said the Attorney General had “hired an out-of-state” private law firm for a lawsuit that, in Stone’s words, “ignores the full range of actions we take to combat scams every day.” Engadget reported that the company has acted to shut down the bad actors behind some of those ads, has sued the creators of others and added some user protections for suspicious messages earlier this year.

What the State Asks For

The release lists a declaratory judgment that Meta violated the Deceptive Trade Practices Act, injunctive relief, restitution to affected Rhode Island consumers, civil penalties for each violation, disgorgement of unjust enrichment, and “the appointment of a receiver, auditor, or monitor to ensure Meta’s compliance with relief ordered by the Court.”

The release also says the suit follows an August 2026 multistate settlement with Meta of up to $17.1 billion, under which Rhode Island will receive at least $49.4 million and up to $64.5 million. Engadget reported that the Consumer Federation of America sued Meta over scam ads in April and Santa Clara County did so in May.

Connecting the Coverage

Our Meta selloff piece reported that the Nasdaq fell 2.6% on 22 October, the day the suit was filed, as a batch of high-profile earnings “ignited fresh fears about the AI spending race,” according to Reuters. Our Meta discovery piece reported that a federal magistrate judge ordered Meta on 19 October to identify each person who approved downloads of shadow-library datasets in the Kadrey v.

Meta cases. What’s new: a state is now testing, in court, how Meta’s AI ad tools and its handling of scam ads fit its public statements.

The Structural Read

As we read it, many of the complaint’s numbers are not new disclosures: the $16 billion projection, the 15 billion ads a day and the $7 billion figure are all attributed to Reuters reporting on Meta documents.

As we read it, the verification figures are the core of the cost argument: the complaint says Meta estimated verification would cost $2 billion, just over 1% of its advertising revenue, and that its program is voluntary and fee-based.

As we read it, the remedy is the part with operating consequences: beyond penalties and restitution, the release asks for a receiver, auditor, or monitor to ensure Meta’s compliance with relief ordered by the Court.

Three Implications

FOR ADVERTISERS The complaint says Meta’s verification program is voluntary and fee-based and that an account needs an email or phone number, a name and a birth date; the release asks the court for injunctive relief.

FOR META The release lists civil penalties for each violation of the Deceptive Trade Practices Act, restitution and a court-appointed monitor among the relief sought.

FOR AI AD TOOLS The complaint names dynamic creative, Advantage+ and a generative AI text feature as part of how scam ads were made and targeted, so the case tests how far a platform answers for what its tools generate.

The Business Engineer Lens

This story maps onto the Business Engineer framework Inside Anthropic’s Permission Layer.

The analysis describes the governance layer this way: “It is where decisions are made about who gets access to frontier intelligence, under what conditions, and for which purposes.”

As we read it, the complaint puts Meta’s ad tools in that frame: it says an advertiser can open an account with an email or phone number, a name and a birth date, and then use generative AI features to make ad variations. The analysis was published in June 2026 about Anthropic; it does not discuss Meta or advertising.

What Is Not Established

These are allegations; a court has not ruled on them. Many of the complaint’s figures come from Reuters reporting on Meta’s internal documents, not from new disclosures. The passages in which the complaint quotes Meta’s internal comparisons with Google are blacked out in the public version we read; a post on X summarising the suit quotes one such line, and we could not check it against that version.

We read the Attorney General’s release, the public complaint and Engadget’s report; we did not read Reuters’ story on the suit and did not contact Meta. Nothing here is investment advice.

The Bottom Line

Rhode Island has sued Meta under its Deceptive Trade Practices Act, alleging the company helped create, target and profit from scam ads while saying it fights them, and asks for penalties, restitution and a court-appointed monitor. Meta told Reuters the suit ignores the actions it takes against scams and that it has removed more than 159 million scam ads.