Anthropic sued over Claude Max subscription limits — customers allege $200 plan doesn't deliver what it promises

A Washington, D.C. customer filed a federal class-action lawsuit against Anthropic on June 15, alleging the company misled subscribers about what its two premium Claude subscription tiers actually deliver — a consumer case that arrives as the company fights a government-imposed model shutdown and prepares for an IPO.1

The complaint, filed in the Northern District of California by plaintiff Karl Kahn, targets Anthropic's two Max subscription tiers: Claude Max 5x at $100/month and Claude Max 20x at $200/month. Both plans are marketed as providing five times and 20 times the computing access of the base Claude Pro plan, which costs $17–$20 per month.

The lawsuit claims the actual usage provided is "far below the advertised amount" and seeks a finding that Anthropic's marketing of the plans is fraudulent. It also requests reimbursement on behalf of all individuals who purchased either Max plan since April 2024.2

Kahn says he moved to the Max 20x tier in April after relying on Claude for heavy coding work. A single five-hour work session consumed 15% of his weekly allowance. After that, the caps started cutting off his access — leaving him, as the complaint puts it, "needing either to halt his work, ration his usage, or purchase additional usage to ensure that he could complete his work." Central to the filing are emails Anthropic allegedly sent in July 2025 outlining per-model weekly usage expectations for each pricing tier — which the plaintiff contends contradict what subscribers actually receive.1

Anthropic declined to comment.

Anthropic introduced Max tiers in April 2025 specifically to address complaints from Claude Pro subscribers who were constantly hitting throttling limits. The promise was more headroom for power users. Critics now argue the math doesn't hold: once the usage cap is reached under Max 20x, overage charges reportedly accumulate at roughly 20 times the base rate — a structure that has reportedly generated bills as high as $50,000 in a single month for extreme-use cases, and $250 in usage value consumed in under 90 minutes in documented community examples.3

The complaint is one of the first times subscriber frustration over opaque AI usage limits has reached federal court — a threshold that consumer attorneys say signals AI subscription charges are now embedded enough in household and professional spending to invite scrutiny.2

Anthropic is in an unusually exposed position. The company filed a confidential S-1 on June 1, making an IPO likely within the next few months. The subscription lawsuit adds a consumer fraud allegation to a record that already includes a $1.5 billion copyright class-action settlement over training data, an ongoing dispute with the Pentagon over its national security designation, and the government-forced shutdown of Fable 5 and Mythos 5 this week.

For investors evaluating the S-1, the Max tier complaint carries specific weight: Claude's direct consumer subscriptions are a visible signal of Anthropic's path to retail revenue, distinct from its enterprise and API business. A court finding that the plans are deceptively marketed would require either a pricing overhaul or a disclosure rewrite — both of which complicate the business narrative Anthropic intends to present in its public filing.

Anthropic's enterprise contracts — which include granular per-user spend caps and usage analytics — make the transparency gap in its consumer tiers an especially uncomfortable contrast to present to a federal judge.2