One year after "DeepSeek Moment," the market witnessed the "Claude Impact"

AI startup Anthropic achieved a key breakthrough this week with its enterprise-level product strategy, prompting a reassessment of the AI competitive landscape. Its tools have triggered a chain reaction in global stock markets, with significant market value evaporation in sectors such as software, legal, financial data, and real estate, indicating that the enterprise software market is facing structural reconstruction.

The San Francisco-based company, founded just five years ago, is conducting a financing round at a valuation of approximately $35 billion and plans to initiate an IPO within the year. Since the beginning of 2024, its annualized revenue has surged from about $1 billion to over $9 billion by the end of 2025. According to media reports citing informed sources, the company expects its annualized revenue to exceed $30 billion by the end of 2026 and anticipates achieving breakeven for the first time in 2028, two years ahead of OpenAI.

This week, several industry-specific tools released by Anthropic have continued to shake the market. Following the launch of its latest flagship model, Claude Opus 4.6, the stock prices of software companies such as Salesforce and Intuit came under pressure again. On Friday, Goldman Sachs announced a partnership to develop an AI agent for automated banking services, further validating the effectiveness of its enterprise path.

Dean Ball, a senior researcher at the Foundation for American Innovation, pointed out that the rapid adoption of Anthropic's model "is the most influential event in the AI field since the launch of ChatGPT." Investors are betting that its tools can systematically transform white-collar workflows, targeting not traditional IT budgets but rather larger workforce expenditures. This marks a shift in AI competition from the technical demonstration phase to a substantive replacement phase for corporate operational cost structures.

Enterprise Market Strategy Proves Effective

Anthropic has chosen a commercialization path that is significantly different from OpenAI, Google, and Meta, focusing not on consumer-facing products but positioning its AI models as tool products for developers and enterprises. This strategy has been strongly validated by market performance over the past week.

According to data from expense management startup Ramp, Anthropic has dominated the AI model API spending market since January, with a share close to 80%. This data reflects the actual spending of users calling AI models through third-party services. Although previous surveys indicated that OpenAI leads in the number of enterprise users, Anthropic's market appeal on the enterprise side is rapidly increasing.

David Hsu, CEO of startup Retool, which helps companies build and manage internal AI tools, noted:

“Anthropic has always focused on the enterprise market, and they found that the key to everything lies in ‘coding.’”

Matt Murphy, a partner at Menlo Ventures, revealed that the Claude Code tool was initially for internal use at Anthropic, and after validating its significant effects, the company quickly transformed it into an external product. This week, Anthropic further released a series of specialized "plugin" tools for vertical industries such as legal, sales, finance, marketing, and customer support, continuously deepening its enterprise service ecosystem

Coding Ability Becomes Core Competitiveness

The software engineering tool Claude Code launched by Anthropic last year has become a leader in the field. This system can read existing codebases of enterprises, autonomously plan and execute tasks, marking the preliminary realization of AI "agent" capabilities. Investors expect this to open up vast new markets, as AI models begin to possess the ability to independently handle complex tasks.

The tool has attracted a large number of developers and has given rise to a new group known as "Claude benders," referring to those who use this tool for high-intensity, marathon-style application development. Early versions have amazed software engineers, with some users stating that the latest model can compress projects that would typically take years into just a few weeks.

Its application range is expanding beyond programming to non-technical positions and tech enthusiasts, with users highly praising its ability to control computers, use browsers, and complete tasks beyond coding.

Despite facing fierce competition from Google and OpenAI (the latter of which just released an upgraded version of the Codex coding tool this week), some OpenAI supporters point out that "the coding market is not equivalent to the enterprise market; it essentially remains in the developer domain." However, investors participating in this round of financing for Anthropic, including Nvidia, Microsoft, and top venture capital firms like Lightspeed, Sequoia Capital, and Altimeter Capital, are betting that the company's tools will go beyond code generation and deeply reshape white-collar workflows.

Returns of a Safety-First Strategy

Anthropic was founded in 2021 by a former OpenAI research team, with co-founders including CEO Dario Amodei and his sister, President Daniela Amodei. Dario Amodei previously served as a researcher at Google and left due to disagreements with OpenAI CEO Sam Altman.

The company has consistently shaped and maintained a cautious, safety-focused public image, with its CEO repeatedly warning through lengthy blog posts about the risks that unrestrained AI development may pose, further reinforcing this characteristic. In 2022, out of concerns about a potential AI arms race, Anthropic proactively delayed the public release of its AI models. In November of the same year, OpenAI launched ChatGPT, which quickly ignited the market, while Anthropic did not follow up with product releases until several months later.

For quite some time, this delay strategy led to Anthropic being viewed as permanently lagging behind OpenAI. However, its long-term investment in safety has played a key role in the recent market recovery.

Anthropic pioneered the "AI feedback-based reinforcement learning" technology, aimed at ensuring its models avoid harmful or unsafe outputs. Unlike ChatGPT, which relies on human feedback regarding the quality and safety of answers, Anthropic employs AI systems to review AI-generated content, with humans only providing guiding principles. Company executives state that this approach not only helps reduce human bias but also significantly enhances the efficiency of model iteration Twelve investors interviewed by the Financial Times pointed out that Anthropic's appeal in the enterprise market, clear product focus, and stable management team make it increasingly viewed as a more robust long-term investment choice than OpenAI. Notably, all seven co-founders of Anthropic are still with the company, while since its establishment in 2015, eight of the eleven founding members of OpenAI have left, and its CEO Sam Altman was once ousted by the board in 2023.

The Advertising Dispute Highlights Intensified Competition

Anthropic publicly committed this week not to introduce advertising in its products, differentiating itself from competitors like OpenAI, which has tested ads in ChatGPT. The company plans to reinforce this stance with a series of satirical ads aired during the Super Bowl, quoting Dr. Dre's lyrics: "What’s the difference between you and me? You speak well, but you don’t do what you say."

Anthropic President Daniela Amodei stated that the ads are not aimed at any specific company, but OpenAI CEO Sam Altman directly responded on X, accusing them of being "clearly dishonest" and "consistent with Anthropic's usual double standards." He later downplayed the controversy in the tech podcast TBPN, calling it a "minor episode" and emphasized that "model capabilities, product progress, and the industrial wave around Codex are what matter."

Despite warnings from industry leaders like NVIDIA CEO Jensen Huang that market reactions may be excessive, pointing out that actual platform building is far more complex than it appears, tech giants such as Microsoft, Amazon, Meta, Oracle, and Google still plan to invest over $600 billion in capital expenditures by 2026, a scale close to Japan's national budget for that year, exceeding Germany and Mexico. The multi-scenario applications of Anthropic and its tools are one of the key factors driving such expenditures.

Neeraj Agrawal, a general partner at Battery Ventures, noted that this year companies may evaluate AI investments with a more cautious attitude, making the outcome difficult to predict. Notably, the firm has not invested in Anthropic or OpenAI. He stated:

"We are at the peak of AI experimentation."

Currently, venture capital firms, corporate executives, and AI researchers generally believe that the competition is far from over, with the rapid iteration of AI technology meaning that current leaders could be surpassed in the short term