Anthropic says an AI security startup's logo looks too much like its own

Anthropic is suing AI-enabled security firm Abnormal AI for trademark infringement, alleging that one of the startup's rebranded logos resembles its own.

Why it matters: The case is an unusual move for Anthropic given Abnormal is also a customer.

• Abnormal builds AI-powered cybersecurity software and Anthropic has lately been expanding into the cybersecurity market.

State of play: Anthropic filed a lawsuit on July 1 against Abnormal AI for trademark infringement.

• The security startup rebranded last year from Abnormal Security to Abnormal AI.
• Anthropic argues in the complaint that the new "A\" version of Abnormal's logo is too similar to its own and that the logo is likely sparking confusion among AI and cybersecurity customers.
• In response, Abnormal AI CEO Evan Reiser wrote in a blog post on July 7 that the company designed its logo in April 2021, "just months after Anthropic was founded."

The other side: Anthropic told Axios its primary concern with the lawsuit is avoiding customer confusion and that it wants Abnormal to remove the "\" from its smaller logo.

• The company told Axios it viewed litigation as a last resort, saying that trademark owners risk weakening their rights if they fail to challenge branding they believe is confusingly similar.

Threat level: The lawsuit seeks Abnormal AI's profits attributable to the allegedly infringing logo, along with damages, court costs and attorneys' fees.

• Anthropic is also asking the court to bar Abnormal from continuing to use the disputed logo and related branding. Anthropic says it is not challenging Abnormal's full wordmark or branding that predated the company's 2025 rebrand.
• Anthropic also told Axios that the lawsuit is about protecting its trademarks, not going after Abnormal's business, and that its own trademark rights date back to 2021.

What they're saying: "Anthropic's case looks pretty strong based on the similarity of the logo and the fact they're in the same field," technology lawyer Denise Howell told Axios.

• In his blog post Reiser argues that Abnormal is in a different market than Anthropic. "Anthropic builds general-purpose AI language models. Abnormal builds specialized behavioral AI models to understand enterprise behavior and stop cyber attacks," he writes.
• "Trademark law is all about preventing confusion, and the chances of confusion seem pretty high here," Howell says, adding that Abnormal may have some arguments to make about the generic natures of letters and the term "AI."

Yes, but: Reiser told Axios in a statement that "the lawsuit's demands go well beyond a design fix."

Zoom in: Anthropic said in its filing that it reached out to Abnormal shortly after its redesign last year, asking the company to change its logo. In the filing, Anthropic added it later sent a formal demand letter in November and that, in April, Abnormal again declined to change its logo.

• Anthropic said in its complaint that its account team spoke with Reiser during the first week of June about their concerns.
• Reiser said in the blog post that while there were prior conversations about the logo before the lawsuit and he was not warned that the AI giant was considering formal legal action.
• "What surprised us is that despite those conversations, we learned about the lawsuit itself from a reporter, not from Anthropic directly," Reiser told Axios in a statement.

Zoom in: Reiser also told Axios in an interview that the lawsuit runs counter to how cybersecurity vendors work together.

• Abnormal is an Anthropic customer, and Reiser estimates Abnormal will spend more than $10 million on their Anthropic account this year.
• "Cybersecurity is just a trust game," Reiser said. "We need customers to trust you to tell us about the problems, they trust you with the data — trust is a really big deal in cyber."

The big picture: Reiser says he views the lawsuit as distracting from his biggest priority: using AI to stop increasingly sophisticated cybercrime.

• "We're trying to go stop crime," he said. "If we spend all the resources, time, and energy arguing with other people about different lawsuits, that's not good for us, it's not good for our customers, and it's not good for actually trying to save the world."

Between the lines: Much of Anthropic's complaint focuses on documenting its cybersecurity work, noting it has spent years building a public presence in the market through research, conferences, government partnerships and newer cybersecurity products.

• Reiser said he has never heard of an Abnormal customer confusing the company with Anthropic, adding that Abnormal sells to "very sophisticated buyers" who run extensive trials before purchasing.
• Abnormal's platform specializes in tracking and detecting malicious activity on a system, whereas Anthropic's Claude offerings are broader and more general-purpose, he added.

What we're watching: Abnormal has yet to formally reply to Anthropic's complaint in the district court.