Class action claims Anthropic misled Claude Max subscribers about usage limits

Claude subscribers have filed an expanded class action lawsuit accusing Anthropic of misrepresenting what its most expensive subscription tier delivers. According to The Verge, the complaint — filed on September 8 by attorneys Monica Vaca and Kati Daffan, both of whom previously worked at the Federal Trade Commission under Lina Khan — argues that marketing for the Claude Max plan led customers to believe they were buying far more usage than they actually received.

What the lawsuit alleges

Max sits above Anthropic's $20-per-month Pro plan and comes in two options: $100 per month for '5x' Pro's usage or $200 per month for '20x'. The suit contends those headline multipliers are hollowed out by fine print. The expanded usage applies to five-hour windows that are themselves governed by a weekly cap, which plaintiffs say results in a total increase in usable capacity far smaller than the marketing graphics imply.

Working out the real terms, the complaint argues, required clicking through two separate hyperlinks just to encounter the word 'session', and then visiting the Pro plan's webpage to find out what a session actually meant. 'You've got to dive deep,' Vaca told The Verge.

Vaca said subscribers often upgrade on the fly after hitting limits in the middle of a project, only to discover the additional headroom is less than expected. Consumers have no way to audit what they are buying, she argued: 'There's no way for them to know exactly what it is they're going to get.' Complaints have circulated online for months, with one Reddit poster comparing the arrangement to handing someone a bigger gas tank while restricting the fuel pump to one gallon every five hours.

Anthropic's defense

The case has already been through one round. An initial complaint was filed in July, then withdrawn and refiled as an expanded class action. In a motion to dismiss the earlier version, Anthropic did not dispute that session limits existed, but argued it had not omitted them — the clarifying information was available to anyone willing to click hyperlinks during the purchase process, which the company likened to turning a product over to read the back label. The Verge reports Anthropic did not respond to a request for comment.

Vaca rejected that framing. Customers cannot inspect what sits inside a service's 'black box', she said, and should not have to hunt through linked pages for qualifications that contradict the headline claim. 'It's a buyer beware approach,' she argued. 'And that's really not fair to people.' She and Daffan spent a combined 38 years at the FTC, and Vaca pointed to established legal ground: 'There is a long line of precedent on false advertising. It's commercial speech. You can't lie when you're marketing a product.'

Rising cost pressure on AI users

The dispute unfolded against a backdrop of intensifying monetisation across the industry. Anthropic has described heavy users as central to its business and has prioritised them even when doing so meant cutting off popular third-party applications such as OpenClaw, The Verge notes. The Max plan launched in April 2025, but the contested weekly limits were not introduced until a few months later, in August 2025, as Anthropic pushed to compete with OpenAI. Over the same period, customers have repeatedly accused leading AI labs of passing their steep operating costs downstream as pressure to turn a profit has grown. In its latest model release, Fable 5.1, Anthropic signalled awareness of the frustration, saying it was 'addressing the feedback we've received from customers on price'.

Vaca added that the case felt compelling for another reason: she and Daffan kept hearing from people who believed they had to pay premium subscription prices to avoid becoming irrelevant in the job market, only to feel shortchanged on services costing $100 to $200 per month.

Why it matters

This is a rare attempt to apply consumer-protection and false-advertising law to AI subscription pricing, and it tests whether rules written for conventional products extend to opaque, metered AI usage. Multiplier-based tiers like Max are a core revenue engine for frontier labs, and similar marketing is common across the industry; if these claims survive a motion to dismiss, subscription advertising well beyond Anthropic could come under scrutiny. The case also gives legal form to a widely shared grievance — that as compute costs climb, AI pricing is becoming less transparent rather than more — and it could force clearer disclosure of session windows and weekly caps. With two former FTC enforcement attorneys leading the charge, AI subscription pricing has landed squarely on the consumer-protection radar.