A federal magistrate judge in California has let a proposed class action over Anthropic’s Claude Max marketing move forward, finding that “The advertising in question is plausibly misleading” (order, Dkt. 50).
U.S. Magistrate Judge Alex G. Tse ruled on November 6, 2026 in Kahn v. Anthropic PBC, case 3:26-cv-05763 in the Northern District of California (docket). The plaintiffs, Karl Kahn and Solomon Smith, allege that the $100 Max 5x and $200 Max 20x plans deliver far less than the “5x or 20x more usage” Anthropic advertises over a full week. Courthouse News reported the ruling (report).
Business Pill · Rate Limits
A short Business Engineer explainer on rate limits: why an AI product caps how much each user can do in a period. It teaches the general idea only and makes no claim about this case.
The key insight: A federal magistrate judge found Anthropic’s ‘5x or 20x more usage’ Claude Max advertising plausibly misleading and let negligent misrepresentation, breach of contract and one consumer-protection claim proceed.
On claude.com, the order notes, Anthropic promises Max 5x subscribers “5x more usage than Pro” and Max 20x subscribers “20x more usage than Pro”. The plaintiffs say those multiples hold per five-hour session but not per week.
According to the order, the complaint calculates that for Claude’s Sonnet 4 model, Max 5x users get 3.5x more usage than Pro users per week and Max 20x users get 6x. Those weekly figures are the plaintiffs’ own calculation in their complaint, not a finding by the court.
The order says that because weekly limits are higher for Max but not 5x or 20x higher, “a Max subscriber may hit his weekly usage limit sooner” than a Pro subscriber will. It quotes one customer complaint: “I am only on my third day of the week”.
Claude Max usage vs Pro: the multiple Anthropic advertises (5x, 20x) against the weekly multiple the plaintiffs calculate for Sonnet 4 (3.5x, 6x), as described in the court’s 6 November 2026 order. The weekly figures are the plaintiffs’ allegation, not a court finding; the percentages are our arithmetic.
What Was Posted on X
‼️ A US judge says Anthropic must face claims that it misleadingly advertised its $ 200-a-month Claude Max plan as “20x more usage than Pro” when it delivers about 6x a week.
The multiplier applies per five-hour session, but weekly caps don’t scale with it. A Max subscriber can run out for the week before a Pro user does, the order notes.
By the plaintiffs’ maths on Sonnet 4, the $100 Max 5x plan gets 3.5x Pro’s weekly usage.
Judge Alex Tse found the ads plausibly misleading. The “per session” fine print sat two hyperlinks away from the sales pages.
— International Cyber Digest (@IntCyberDigest) Nov 07 23:45
International Cyber Digest posted on X that a US judge says Anthropic must face claims over how it advertised the $200-a-month Claude Max plan as “20x more usage than Pro”. The post gives about 6x a week for Max 20x, and 3.5x for the $100 Max 5x plan on Sonnet 4 by the plaintiffs’ maths. The court’s order, which this piece is built on, describes both weekly figures as the plaintiffs’ calculation.
What the Judge Decided
Anthropic argued that customers should read “more usage” as more usage per session, pointing to a help page. Judge Tse wrote that “Anthropic’s per session disclosure is located two hyperlinks away from the primary sales pages on claude.com”.
The order says the disclosure sits “outside the sales funnel” and compares it to a back-label disclaimer on a shampoo bottle. Because payments are due monthly, it finds that reasonable consumers would plausibly read the multiples as applying per month.
The court denied the motion to dismiss on Smith’s claim under California’s Consumers Legal Remedies Act, the negligent misrepresentation claim and the breach of contract claim. It dismissed Kahn’s Consumers Legal Remedies Act claim, the unfair competition and false advertising law claims, and the implied covenant claim, all with leave to amend.
On the implied covenant claim, the order says “Anthropic didn’t tighten usage limits as a bait-and-switch”. It sets December 2, 2026 for an amended complaint; if the plaintiffs do not amend, Anthropic must answer by December 16, 2026.
Anthropic’s Response
Courthouse News reported that representatives for Anthropic and the plaintiffs did not immediately respond to a request for comment. We found no public statement from Anthropic on the ruling.
How Session and Weekly Limits Differ
On October 30, Anthropic renamed Max 20x as Ultra and removed the 5-hour limit for verified Ultra accounts, saying weekly limits were unchanged (our piece). The lawsuit concerns how the plans were advertised; the order does not address the rename.
Using the complaint’s own Sonnet 4 figures, as described in the order, the 6x weekly multiple for Max 20x is 30% of the advertised 20x, and the 3.5x weekly multiple for Max 5x is 70% of the advertised 5x. That is our arithmetic on the plaintiffs’ calculation, not a court finding.
The Structural Read
As we read the order, the dispute turns on the unit of the promise: the multiples hold per five-hour session, while the plaintiffs say weekly limits do not scale the same way.
As we read it, the court did not test the plaintiffs’ weekly figures; the 3.5x and 6x numbers come from the complaint and remain allegations.
By our arithmetic on the plaintiffs’ Sonnet 4 figures, the 6x weekly multiple is 30% of the advertised 20x, and 3.5x is 70% of the advertised 5x.
Magistrate Judge Alex G. Tse, order of 6 November 2026
“Anthropic’s per session disclosure is located two hyperlinks away from the primary sales pages on claude.com”
Three Implications
FOR SUBSCRIBERS The order says that because payments are due monthly, reasonable consumers would plausibly read the 5x and 20x multiples as applying per month.
FOR ANTHROPIC The order dismissed the implied covenant claim, saying Anthropic didn’t tighten usage limits as a bait-and-switch, and gave the plaintiffs until December 2, 2026 to amend.
FOR AI PLAN PRICING The case concerns how a usage multiple is advertised when a plan has both per-session and weekly limits.
As we read it, the lawsuit is about how one of those bundled terms, the usage multiple, is described on the sales page. The framework does not discuss this case.
What Is Not Established
The ruling is on a motion to dismiss, which tests whether the claims are plausibly pleaded. The court has not found that Anthropic misled anyone, has not certified a class, and has not tested the plaintiffs’ weekly usage figures. This piece makes no forecast about the outcome.
This Week In The AI Supercycle
The Great Convergence of Capital, Power, and Geopolitics
This week’s issue: for two years the AI build-out was priced as a supply story, and last week as a credit story. This week it was priced as a permission story, and the answer arrived as a cost, not a ban.
Judge Alex G. Tse found Anthropic’s “5x or 20x more usage” Claude Max advertising plausibly misleading and let negligent misrepresentation, breach of contract and one consumer-protection claim proceed. The plaintiffs allege the plans give 3.5x and 6x Pro’s weekly usage on Sonnet 4. An amended complaint is due December 2, 2026.
A note on sourcing. The facts and quotes are from the court’s 6 November 2026 order in Kahn v. Anthropic PBC (Dkt. 50) and the Courthouse News report of the same day. The weekly usage multiples are the plaintiffs’ calculation as described in the order. The percentages are our arithmetic. This is not legal or investment advice.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
