The FTC made DoNotPay answer for its robot lawyer claims

The conversation

On 25 September 2024, the Federal Trade Commission announced a sweep of enforcement actions called Operation AI Comply, one of which targeted DoNotPay, a company that had marketed its subscription service as ‘the world's first robot lawyer.’ According to the FTC's own complaint, as the press release describes it, DoNotPay had told consumers its AI chatbot could ‘sue for assault without a lawyer,’ generate ‘perfectly valid legal documents in no time,’ and eventually ‘replace the $200-billion-dollar legal industry with artificial intelligence.’

What the documents show

The FTC's release states that, according to the complaint, DoNotPay did not test whether its chatbot's output matched the standard of a human lawyer and did not hire or retain any attorneys to check the quality of its law-related features. A second product claim, that the service could scan a small business's website for legal violations that might otherwise cost the business up to 125,000 dollars, is also described in the complaint as ineffective. The company agreed to a proposed Commission order, authorized by a 5-0 vote, requiring a 193,000 dollar payment and a notice to consumers who subscribed between 2021 and 2023 about the limits of the service's law-related features, plus a bar on advertising that the service can substitute for a professional without evidence to back that claim. The FTC's later finalization release, dated 11 February 2025, states that the Commission finalized this order after receiving five public comments, on a further 5-0 vote taken 16 January 2025.

The system boundary

The FTC's own materials frame the case around the gap between a marketing claim and the testing behind it, not around a general finding that the underlying chatbot technology cannot work: the agency's language centers on DoNotPay's failure to test its output against a human-lawyer standard and its decision not to retain attorneys, which the release treats as the missing human check behind an automated legal-document and advice service. The order's forward-looking term, barring professional-substitute claims made without supporting evidence, defines the boundary going forward as evidence-backed marketing rather than a ban on the product category.

Where it fails

The available FTC documents describe allegations and a negotiated settlement, not a litigated finding on the merits, and the release does not state that DoNotPay's chatbot never produced any accurate output, only that the company had not tested or staffed for that standard. A builder marketing an AI tool as a substitute for licensed professional work should read this case as being about the evidentiary basis for that specific comparison.

• Is there documented testing behind any claim that an AI tool performs at the level of a licensed professional it is compared to?
• Does the marketing distinguish between what the documented testing supports and a broader claim of professional-equivalent performance?
• If past customers were affected by an unsupported claim, is there a notice process comparable to the one this order required?

Nothing in the FTC's own releases cited here supports describing DoNotPay's technology as having caused any specific consumer's legal loss; the agency's stated case is about the adequacy of the company's claims and testing.