Anthropic faces lawsuit alleging false advertising over Claude Max subscription usage limits

Anthropic Lawsuit Alleges Misleading Claude AI Pricing

AI company Anthropic is facing a proposed class-action lawsuit against Anthropic that accuses it of false advertising over its premium Claude Max subscription plans. Filed on June 14 in the U.S. District Court for the Northern District of California, the complaint alleges that the actual Claude Max usage limits fall dramatically short of what the company advertises to paying customers1

1

.

The plaintiff, Karl Kahn, a Washington D.C.-based user, claims he upgraded his subscription twice before reaching the Max 20x plan in April 2025, seeking expanded access for coding work. What he encountered instead were frequent usage caps that bore little resemblance to the promised multipliers2

2

. The lawsuit seeks class-action status on behalf of all U.S. consumers who purchased Max 5x or Max 20x subscriptions since April 2025, with damages exceeding $5 million1

1

.

Oversold Claude Plans Deliver Less Than Promised

At the heart of the case are Anthropic's two premium tiers: Max 5x, which costs $100 per month and promises five times the usage of the Pro plan, and Max 20x, priced at $200 per month with a pledge of 20 times Pro's capacity. The Claude Pro subscription tier costs between $17 and $20 per month1

1

.

According to the complaint, the Max 20x plan delivers "just six to eight times the usage of Pro" rather than the advertised 20 times, while Max 5x provides "just three-and-a-half times the usage of Pro" instead of five times1

1

. These oversold Claude plans have left subscribers frustrated and feeling misled about what they purchased.

Kahn's experience illustrates the problem. During a single five-hour coding session, he burned through approximately 15 to 20 percent of his weekly data allocation, quickly hitting rate limits that forced him to either halt work, ration prompts, or purchase additional credits2

2

4

4

.

Transparency in AI Pricing Under Scrutiny

The Anthropic lawsuit highlights a critical issue with transparency in AI pricing. According to the filing, Anthropic's website "is a black box, without any meaningful description of how usage is calculated"1

1

. The company doesn't clearly define what constitutes a single session or explain exactly how many prompts each plan allows during relevant time windows.

Kati Daffan, founding partner at Vaca Daffan LLP and representing Kahn, told Decrypt that "it's really not easy for a normal consumer to know if they're getting the amount they were promised, or if they're not, because that information simply isn't provided"5

5

. This opacity may explain why similar lawsuits hadn't previously emerged, despite widespread complaints on platforms like Reddit where users described the premium plans as "extremely misleading"1

1

.

The lawsuit alleges that emails Anthropic sent to Max subscribers in July 2025 outlined weekly token allowances that were "far below the advertised amount of usage"3

3

. All large language models operate on a currency known as tokens, which represent words, character groups, and punctuation converted into numbers. Both input and output costs can vary greatly depending on prompt complexity2

2

.

AI Subscription Models Face Growing Scrutiny

This case arrives as AI subscription models face mounting pressure over costs and value. Companies like Amazon and Uber have recently limited employee use of AI tools citing high costs, while OpenAI CEO Sam Altman has acknowledged AI inference costs as "a huge issue" for enterprise customers3

3

.

The lawsuit filed against Anthropic is rooted in consumer protection laws that prohibit companies from misleading users about products and services. "This is a really good type of case for a class action, because the law is very clear that companies have to be honest about their advertising and marketing," Daffan explained5

5

.

The complaint seeks damages, restitution, injunctive relief, and attorneys' fees5

5

. It highlights a disconnect between how Anthropic treats enterprise versus consumer customers: enterprise plans include granular spend caps and usage analytics, while consumer tiers remain opaque4

4

.

Rising Costs and Agentic Models Amplify Concerns

The timing of the Anthropic lawsuit coincides with the recent launch of Claude Fable 5, a model touted for its agentic capabilities that allow it to perform complex, autonomous tasks. Shortly after its debut, users complained about burning through their usage windows far faster than with earlier models3

3

.

More agentic models consume significantly more tokens. While rewriting an email uses minimal token allowances, designing a website landing page could require multiple steps, deployment of sub-agents, and substantially more computing power3

3

.

The lawsuit underscores a fundamental tension: traditional software subscription models don't map neatly to how much compute large language models require to work at scale. Right now, venture capitalists are offsetting those costs, but the problem will likely intensify when Anthropic and OpenAI go public2

2

.

Competitors have already seized on affordability concerns. During Microsoft's recent Build conference, the company unveiled MAI-Thinking-1, which performs comparably to Claude Opus 4.6 at lower cost, with AI lead Mustafa Suleyman noting that "many people are urgently looking for alternatives" to Anthropic's expensive models3

3

. OpenAI is also reportedly considering significant cuts to current token prices3

3

.

Anthropic declined to comment on the lawsuit2

2

. The allegations remain unproven, and no class has been certified yet4

4

. However, the case raises questions every AI company will need to answer as subscription costs become a routine household expense: what exactly does your plan buy, and how can users verify they're getting what they paid for?