Anthropic has told investors something most companies would rather not admit. A meaningful slice of its 2026 revenue came from customers who were breaking its rules.
The company behind the Claude models estimates that terms-of-service violators produced 5% to 7% of its revenue across the first nine months of 2026. Most of those violators, according to Anthropic, were Chinese AI labs using Claude to train their own competing models.
How the distillation problem worked
The practice at the center of this is called model distillation. In AI terms, one lab sends huge volumes of prompts to a stronger model. It then uses the responses as training data for its own system. The result is a cheaper imitation built on someone else’s research spending.
Anthropic says the activity ran through thousands of fake accounts. Those accounts generated millions of interactions with Claude to harvest training data in violation of its terms.
The scale was not subtle. In February 2026, labs including DeepSeek and Moonshot used around 24,000 fake accounts to carry out more than 16 million interactions with Claude, according to the research findings.
Here was the uncomfortable part. Those fake accounts were still paying for access, which means the copying showed up on Anthropic’s income statement as real revenue.
The crackdown and its cost
Anthropic responded with a set of tighter security controls. These included access restrictions and bans on tens of thousands of accounts.
The company says it materially cut off the violating activity in the third quarter of 2026. It also deployed enhanced security measures in October 2026, which it claims stopped 90% to 99% of unauthorized data extraction.
Third-quarter revenue came in at $16.8 billion, up from $11.5 billion in the second quarter. The research describes that as roughly 45% quarter-over-quarter growth.
For context, Anthropic’s total revenue for the entire first half of 2026 was approximately $16.3 billion.
A valuation under discussion
Anthropic initially aimed for a valuation of around $2 trillion, according to the research findings.
Investor discussions have since suggested that figure could potentially be cut to $1.5 trillion, tied to the slowing growth rate.
Anthropic has framed the violations as more than a contract dispute. It has described distillation as a threat to its intellectual property and to national security, a framing that puts the issue alongside broader US concerns about AI competition with China.
Concentration risk in the draft S-1
The rule-breaking revenue is not the only exposure Anthropic has had to disclose. A draft S-1 filing outlines heavy customer concentration as a key risk.
According to that draft, 20% of Anthropic’s first-half revenue came from just two large clients. Neither was named.
The filing also points to significant long-term commitments for computing resources. Those obligations could affect the company’s future financial stability if revenue growth does not keep pace.
Disclosure: This article was edited by Diego Almada Lopez. For more information on how we create and review content, see our Editorial Policy.