A cloud customer has sued Anthropic, alleging the company misled consumers about the usage limits of its Max plan. Per wall street journalWashington, D.C.-based resident Carl Kahn filed a federal lawsuit on Monday accusing Anthropic of limiting how much of the cloud he can use. The lawsuit claims that the actual usage limits of Anthropic’s Max 5x and Max 20x tiers are difficult to determine and appear to be lower than the limits advertised by the company on its website and elsewhere.
Anthropic offers three different payment tiers for individuals. The company’s most affordable plan, Cloud Pro, starts at $17 per month and offers “at least five times the usage per session compared to our free service” during peak hours. Anthropic writes on its website, “The number of messages you can send will vary depending on the length of the message, including the length of the files you have attached, the length of your current conversation, and the model or feature you are using.” From April 2025, the company has also introduced Max 5 and 20x plans, which promise five and 20 times higher usage limits than the Pro package. Both plans cost $100 and $200 per month respectively.
“The actual usage provided by the Max 5x and Max 20x plans is significantly less than the amount of usage advertised,” the lawsuit alleges. After starting to use Cloud Code, Khan upgraded to the Max 20x plan in April this year, but soon found that his weekly usage limit was expiring. One five-hour session cost him 15 percent of his weekly allowance. The lawsuit is seeking class action-status on behalf of other US consumers who purchased Max subscriptions since Anthropic began offering Max subscriptions last year.
Anthropic declined to comment.
Anthropic’s rate limits are a frequent topic of discussion on Reddit, with one user recently complaining about how they exceeded their five-hour limit after just one cloud code prompt. Last July, the company imposed weekly rate limits on cloud code usage in response to users who were running the coding agent “persistently in the background, 24/7.”
All major language models work on a currency called a token. When you type a question into the cloud’s prompt bar, it converts words, groups of characters, and punctuation marks into numbers that represent various patterns and relationships the anthropic model learned in training. For a model to generate a response, there are both input and output costs, and they can vary significantly depending on the complexity of the prompt.
The lawsuit highlights a long-growing gap between consumer expectations and the actual cost of providing AI inference. Traditional software subscription models do not clearly address how much computation an LLM requires to work at scale. Right now, venture capitalists are covering those costs, and if anything, the problem is likely to become more severe when Anthropic and OpenAI go public.
<a href