4. Anthropic

Founders: Dario Amodei (CEO), Daniela Amodei, Chris Olah, Jack Clark, Jared Kaplan, Sam McCandlish, Tom Brown
Launched: 2021
Headquarters: San Francisco
Funding: N/A
Valuation: $61.5 billion
Key Technologies: Artificial intelligence
Industry: Enterprise technology
Previous appearances on Disruptor 50 list: 1 (No. 7 in 2024)

If OpenAI's explosive growth is best measured by consumer adoption, its rival Anthropic is the gold standard for enterprise use of large language models. 

That is not to say OpenAI isn't making gains in enterprise deals, and Anthropic is popular with millions of individuals for gen AI tasks. But its AI-as-a-service model, in particular for coding tasks where generative AI is making huge leaps in ability, has led the Alphabet- and Amazon-backed startup to exponentially grow its business-to-business deals, and revenue.

Anthropic just hit $3 billion in annualized revenue, an increase in under one year from the $1 billion revenue pace it was on last December, with many customers paying as much as $100,000. The company told CNBC that annual revenue from code generation and software development rose by 10x in the last three months of 2024.

From March to May 2025 alone, annual recurring revenue grew by $1 billion. 

"We've looked at the IPOs of over 200 public software companies, and this growth rate has never happened," Meritech General Partner Alex Clayton recently told CNBC.

The growth has pushed up the firm's valuation, and its access to various forms of capital. In mid-May, it received a $2.5 billion credit line. That followed a $3.5 billion funding round in March that valued the company – which was founded by former OpenAI executives and first debuted its Claude chatbot in 2023 – at $61.5 billion.

Companies that use Anthropic AI include top coding startups, including GitLab and Cursor, as well as big names across the market, from Intuit to Pfizer, Salesforce, Asana, DoorDash and fellow disruptor Canva.

Claude's biggest moment yet came in May, when Anthropic launched two models, called Claude Opus 4 and Claude Sonnet 4, that can work for seven-consecutive hours without performance degradation, a milestone in AI productivity. 

"I do a lot of writing with Claude, and I think prior to Opus 4 and Sonnet 4, I was mostly using the models as a thinking partner, but still doing most of the writing myself," Mike Krieger, Anthropic's chief product officer, recently told CNBC. "And they've crossed this threshold where now most of my writing is actually ... Opus mostly, and it now is unrecognizable from my writing."

Last October, it launched its "agentic AI" rival to offerings from Meta and OpenAI's Operator, called Computer Use, with Claude effectively being able to view a computer monitor, use buttons, type, search the internet and execute real-time tasks. Those types of capabilities are being adopted by companies including PayPal, Square, Plaid, Atlassian and Asana.

Anthropic has been pouring its R&D dollars into more complex tasks for business use cases, shifting away from some focus on B2B and consumer chatbot technology earlier in its history.

Anthropic has not been free of AI controversy — Reddit recently sued the company, alleging unauthorized use of content — and its leadership has not shied away from sensitive AI debates, whether the issue is the threat to human jobs or geopolitics.

Its CEO Dario Amodei recently predicted that AI could wipe out as much as 50% of entry-level office jobs. The company has also taken a vocal stance on the trade war with China and export controls that the U.S. government has placed on chips, after a year marked by the emergence of China's DeepSeek, a direct threat to both Anthropic and OpenAI's lead in the AI race. Anthropic has argued for tighter export controls, referencing DeepSeek's rise, and sparked a war of words with chip giant Nvidia when it claimed in a blog post that Chinese smugglers hide AI chips in "prosthetic baby bumps" and "packed alongside live lobsters."