Artificial intelligence giant Anthropic is facing a newly expanded class action lawsuit from dissatisfied subscribers who allege the company deceptively advertised the usage limits of its highest-tier Claude subscriptions. The legal challenge targets the core of the artificial intelligence industry’s growing monetization squeeze, highlighting a widespread point of frustration among power users who feel shortchanged as top AI labs scramble to offset steep operating costs.
The lawsuit was brought by attorneys Monica Vaca and Kati Daffan, both former officials at the Federal Trade Commission who worked under former FTC Chair Lina Khan. Their combined 38 years of experience at the regulatory agency informs a complaint that marks a rare and notable attempt to legally penalize an AI company for allegedly misleading commercial speech regarding product capabilities and resource allowances.
At the center of the dispute is Anthropic’s "Max" plan for its Claude AI assistant, an expensive upgrade designed for heavy users who routinely push past the boundaries of the standard $20-per-month Pro tier. The Max plan is offered at two distinct pricing levels: $100 per month for what is advertised as "5x" the usage limits of the Pro plan, or $200 per month for a "20x" increase. However, the class action complaint alleges that these prominent marketing figures are highly misleading because they are obscured by complex fine print and restrictive operational structures.
According to the lawsuit, while Anthropic advertises that subscribers will receive five or twenty times the overall usage of the Pro plan, those capacity increases are not distributed evenly across the billing cycle. Instead, the promised capacity is restricted to narrow five-hour chunks of time that are further subjected to strict weekly limits. The plaintiffs argue that this combination results in a vastly smaller total increase in functional usage capacity than what consumers are led to believe when they purchase the upgrade.
To uncover these operational limitations, consumers are forced to navigate a maze of digital disclosures. "You’ve got to dive deep," Monica Vaca said in an interview discussing the user experience. According to Vaca, understanding the actual terms requires a subscriber to click through multiple hyperlinks just to encounter terms like "session," and then navigate to a completely separate webpage detailing the Pro plan to find a definition of what that session actually entails.
Frustration over these terms has been widely voiced within online communities and user forums. One frustrated subscriber wrote on Reddit about their experience with the $100-per-month tier: "The weekly allowance is what the pricing page makes you think you’re buying. The rolling 5-hour window is what actually controls whether you can work… It’s like giving someone a bigger gas tank while keeping the fuel pump limited to one gallon every five hours."
Vaca explained that Claude users frequently upgrade on the fly when they hit abrupt usage walls while in the middle of urgent professional projects. Enticed by the promise of dramatically expanded capacity, they pay the steep monthly fee only to experience immediate disappointment. "People see that they are going to get this dramatically expanded usage… What we hear from people, though, is that when they sign up, they are surprised that they’re not getting the usage that they thought they were getting," Vaca noted.
The lawsuit arrives against a broader backdrop of growing economic pressure across the generative artificial intelligence sector. Over the past eight months and beyond, as investors increasingly demand paths to profitability, consumer-facing AI customers have frequently complained that major labs are subtly or overtly passing down their massive operating expenses onto end-users.
Anthropic originally introduced the Max plan in April 2025, but the company did not impose the contested weekly limits until several months later in August, precisely as it intensified its competitive push against rivals like OpenAI. The financial squeeze on users has become so pronounced that Anthropic even acknowledged cost concerns in the documentation for its latest model release, Fable 5.1, explicitly noting that it was "addressing the feedback we’ve received from customers on price" within the new model’s pricing structure.
Anthropic’s business model has increasingly prioritized power users, even when those strategic decisions involve cutting off other popular applications, such as OpenClaw. Yet, this heavy reliance on high-paying power users risks serious friction when those same customers feel misled about the value they receive.
The legal battle has already seen preliminary procedural maneuvering. The complaint was initially filed in July before being temporarily withdrawn and refiled as an expanded class action. In a motion filed to dismiss the earlier version of the case, Anthropic defended its marketing practices by arguing that it did not omit session limits from its promotional materials. Instead, the company maintained that the clarifying information was technically available to any consumer willing to look for it during the purchase flow.
"Accessing this clarifying information… required nothing more than clicking hyperlinks available in the purchase process—the digital equivalent of flipping a product over to read the back label," Anthropic wrote in its motion to dismiss the initial complaint. The company did not respond to subsequent requests for comment regarding the newly expanded lawsuit.
Vaca strongly disagrees with the notion that the terms are transparent or easy for an ordinary consumer to understand. "This is hard for consumers — they don’t know what’s in that black box," she said, emphasizing the inherent information asymmetry between major technology developers and everyday users. "They have to rely on the claims the marketer is giving. There’s no way for them to audit it. There’s no way for them to know exactly what it is they’re going to get… They’re basically taking a leap, and they’re believing in an honest marketplace."
Furthermore, Vaca pointed out that the emotional and financial stakes are heightened because many subscribers feel a compelling professional necessity to adopt these expensive tools. Daffan and Vaca began hearing from individuals who felt they had no choice but to pay premium subscription costs—ranging from $100 to $200 per month—simply to avoid professional obsolescence in a shifting job market. Realizing they were not receiving the operational capacity they paid for under those high-stakes conditions proved uniquely frustrating for workers.
Drawing on their extensive regulatory backgrounds, the attorneys emphasize that foundational legal protections govern commercial communications, regardless of whether the product is physical merchandise or a cutting-edge software service. "There is a long line of precedent on false advertising. It’s commercial speech. You can’t lie when you’re marketing a product," Vaca said.
Even if Anthropic buries its operational stipulations somewhere within its web architecture, Vaca argues that the burden of discovery should never rest solely on the consumer. "It’s a ‘buyer beware’ approach," she concluded. "And that’s really not fair to people."
Leave a Reply