A Washington, D.C.-based Claude subscriber has filed a federal lawsuit against Anthropic, alleging that the company misled consumers about the usage limits of its Max subscription plans. According to Engadget, the lawsuit, filed by Karl Kahn on Monday, accuses Anthropic of overselling how much he could use Claude. It seeks class-action status on behalf of other U.S. consumers who purchased a Max subscription since the plans launched last year. The case underscores a growing tension between consumer expectations for AI services and the underlying economics of large language model inference — a tension that enterprise procurement teams must now navigate.
The Allegations
Anthropic offers three paid tiers for individual users. The Claude Pro plan costs $17 per month and, according to Anthropic's website, "offers at least five times the usage per session compared to our free service" during peak hours. However, the company also notes that "the number of messages you can send will vary based on message length, including the length of files you attach, the length of your current conversation, and the model or feature you use."
Since April 2025, Anthropic has offered two higher-tier plans: Max 5x for $100 per month and Max 20x for $200 per month. These plans promise up to five and 20 times higher usage caps than the Pro package, respectively. The lawsuit alleges that "the actual usage provided by the Max 5x and Max 20x plans is far below the advertised amount of usage." Kahn upgraded to the Max 20x plan in April 2025 after beginning to use Claude Code. He quickly hit his weekly usage limits, with one five-hour session consuming 15 percent of his weekly allowance.
| Plan | Monthly Price | Advertised Usage Cap | Actual Experience (per lawsuit) |
|---|---|---|---|
| Claude Pro | $17 | At least 5x free usage per session | Variable, depends on message length |
| Max 5x | $100 | Up to 5x Pro | Far below advertised, per lawsuit |
| Max 20x | $200 | Up to 20x Pro | 5-hour session used 15% of weekly limit |
The Pricing Disconnect
Anthropic's rate limits have been a frequent topic of discussion on Reddit, Engadget reported, with one user complaining they blew through a five-hour limit after a single Claude Code prompt. Last July, the company imposed weekly rate limits on Claude Code usage in response to users who had been running the coding agent "continuously in the background, 24/7."
The lawsuit highlights a fundamental mismatch between traditional software subscription models and the cost structure of AI inference. All large language models operate on a currency known as tokens. When a user types a question into Claude's prompt bar, the system converts words, characters, and punctuation into numbers that map to patterns Anthropic's models learned during training. Both input and output have costs, which vary greatly depending on prompt complexity. Traditional software subscriptions assume predictable resource consumption, but AI inference costs are variable and usage-based.
Implications for Enterprise AI Procurement
For enterprise technology leaders, the case serves as a cautionary tale. The lawsuit demonstrates that estimating AI service costs based solely on advertised usage multiples can be misleading. Engadget noted that "venture capitalists are offsetting those costs" currently, but the problem is likely to become more acute when Anthropic and OpenAI go public. Enterprise buyers should demand transparent pricing models that account for variable factors like message length, file attachments, and model complexity. They should also negotiate service-level agreements (SLAs) that clearly define usage limits and provide mechanisms for monitoring consumption in real time.
Anthropic declined to comment on the lawsuit, according to Engadget. As the case progresses, it may set a precedent for how AI companies communicate usage limits — and how enterprises protect themselves from unexpected cost overruns.
Sources:
