Anthropic Is in the AI Business. So Why Is It Selling Safety?

Anthropic has a remarkable pitch: Artificial intelligence could transform civilization, potentially threaten it, and the company best positioned to help manage those risks is…Anthropic!

This is a smart way to differentiate a business in a market where the technology is advancing faster than most users can keep up. But it’s a pretty uncomfortable message coming from one of the companies racing to build increasingly powerful AI. 

And now, Anthropic has effectively acknowledged the commercial logic behind that positioning.

In its latest IPO prospectus, filed November 10 as the company prepared to market its shares to investors, Anthropic describes safety research as central to its mission, but also as a source of customer trust, enterprise adoption and competitive advantage. The company is targeting an initial offering price of $70 to $78 per share.

That is Anthropic explaining its own business model, not an accusation from a skeptical competitor. 

Two things can be true at the same time: Anthropic's researchers genuinely believe advanced AI presents extraordinary risks including biological weapons and advanced cyberattacks. And Anthropic has built a commercial strategy in which being the company that takes those risks most seriously is a competitive advantage.

But here's the paradox: Anthropic believes AI is dangerous enough that it needs to help lead the race to develop it. Staying competitive means building increasingly powerful AI, which introduces new risks and makes its safety mission even more important.

When the solution to dangerous AI is building even more powerful AI, who decides when enough is enough?

In its November 10 amended IPO filing, Anthropic makes an unusually direct connection between its safety mission and its commercial success: "Our commitment to our mission instills confidence in our customers to choose Claude as a trusted partner," the company states.

It explains that investments in alignment, interpretability, cybersecurity and enterprise controls are intended to make Claude predictable and dependable enough for increasingly consequential business applications. 

The filing then describes the commercial payoff: Greater trust allows enterprises to delegate more work to Claude and integrate it more deeply into their operations.

Think about what that means.

Anthropic competes with OpenAI, Google and other AI labs in a market where model capabilities are improving rapidly. Most enterprise buyers can't independently evaluate the underlying architecture or determine which model will lead the next benchmark.

But they can understand the value of reliability.

A company connecting AI to its customer records, internal documents, financial systems and daily operations wants confidence that the technology won't expose sensitive information or create an embarrassing headline in the Wall Street Journal.

Anthropic has made that confidence central to its identity.

The numbers suggest the enterprise strategy is gaining traction. According to its filing, Anthropic had more than 11,900 customers generating at least $100,000 in trailing annual revenue as of September 30, 2026, including more than 1,900 generating at least $1 million.

That growth doesn't prove customers chose Claude specifically because of its safety positioning, but the filing makes clear that Anthropic believes safety and trust help drive adoption.

As TechCrunch reported, Anthropic is also building an ecosystem designed to keep enterprise customers using multiple Claude products. More than half of its customers spending at least $100,000 annually now use both the Claude platform and one of its applications, such as Claude Code or Cowork.

It also describes a self-reinforcing commercial flywheel: More customers generate revenue that funds further research and development, which produces even more capable models, which attract more customers, etc.

Safety is central to this strategy; Anthropic presents it as one of the advantages that helps the entire system work.

None of this means Anthropic is fabricating the dangers it warns about.

As WIRED reported in its examination of Anthropic's IPO prospectus, the company's risk disclosures go well beyond ordinary business concerns. They include the possibility of catastrophic or existential harm, AI-enabled cyberattacks, biological weapons development and the loss of human control over advanced systems.

Researchers disagree about the likelihood of many of the concerns, especially scenarios involving superintelligent systems escaping human control. But some of the strongest warnings are coming from people who have worked directly on the technology, including researchers who have walked away from the companies building it.