For months, the AI chatbot race looked like a two-horse contest. OpenAI’s ChatGPT dominated. Google’s Gemini held a comfortable second position. Everyone else scrambled for scraps.
That calculus changed dramatically in March 2025.
Anthropic’s Claude more than doubled its market share in a single month, climbing from 3.1% in February to 6.9% in March, according to data from web analytics firm Similarweb. The numbers, first reported by MakeUseOf, represent one of the most significant shifts in competitive positioning since the generative AI boom began in late 2022. Claude’s web visits surged to approximately 131 million in March, up from around 58 million in February — a staggering 126% increase that no other major player in the space came close to matching during the same period.
The gains didn’t come from thin air. They came largely at the expense of the incumbents.
ChatGPT’s market share slid from 64.7% to 59.7%. Still dominant, yes, but the five-percentage-point erosion in a single month is the kind of movement that gets noticed in boardrooms. Google’s Gemini dropped from 21.5% to 19.2%. Microsoft’s Copilot fell from 4.4% to 3.6%. Even DeepSeek, the Chinese AI startup that had generated enormous buzz earlier in the year, saw its share retreat from 3.2% to 2.8%.
Claude was the only major chatbot to gain ground. And it didn’t just gain — it surged.
What’s Driving the Shift
The timing of Claude’s breakout isn’t coincidental. Anthropic released Claude 3.7 Sonnet in late February 2025, a model that introduced what the company calls “extended thinking” — a hybrid reasoning capability that allows the AI to work through complex problems more deliberately before generating a response. The feature positions Claude somewhere between a standard conversational chatbot and the more computationally expensive reasoning models that OpenAI has been developing with its o1 and o3 series.
Developers noticed. So did power users.
Claude 3.7 Sonnet earned strong marks in coding benchmarks and complex analytical tasks, areas where professional users are willing to switch tools if they find a meaningful performance edge. Anthropic also made the model available across its free and paid tiers, lowering the barrier for new users to experience the upgrade firsthand. That accessibility matters enormously in a market where most consumers still default to whatever tool they tried first.
But model quality alone doesn’t explain a doubling of market share. Distribution and awareness played critical roles. Anthropic has been steadily expanding Claude’s availability through API partnerships, enterprise agreements, and integrations with platforms like Amazon Web Services, where Claude serves as a flagship model on the Bedrock platform. The company closed a massive funding round in early 2025, reportedly raising $3.5 billion at a valuation near $60 billion, giving it the financial firepower to invest aggressively in growth marketing and infrastructure.
There’s also the matter of competitive stumbles. OpenAI faced a turbulent stretch in early 2025, with public debates over its corporate restructuring, the rollout of GPT-4.5 receiving mixed reviews on cost-effectiveness, and ongoing controversy about the company’s shift from its nonprofit roots. None of these issues were fatal to ChatGPT’s position, but they created openings. When the market leader shows even minor vulnerability, competitors with strong products can capitalize quickly.
Google’s Gemini, meanwhile, has struggled to convert its massive distribution advantage — integration with Search, Android, and Workspace — into the kind of enthusiastic user engagement that drives repeat visits. The product has improved substantially since its rocky Bard-era debut, but it still carries a perception gap among technical users who view it as a tier below the best available models.
The Numbers in Context
A 6.9% market share might not sound like much in absolute terms. But context matters here.
The AI chatbot market is growing rapidly in total size, which means Claude’s percentage gains represent an even larger increase in absolute user engagement. Going from 58 million to 131 million monthly web visits means Anthropic added roughly the equivalent of its entire prior user base in a single month. For a company that was virtually unknown outside of AI research circles two years ago, that trajectory is remarkable.
It’s also worth examining who these users are. Claude has developed a particularly strong following among software developers, researchers, and professional writers — demographics that tend to be vocal about their tool preferences and influential in driving broader adoption within organizations. When a senior engineer at a Fortune 500 company starts using Claude for code review and recommends it to their team, the downstream effects on enterprise procurement decisions can be significant.
The Similarweb data captures only web-based visits and doesn’t fully account for API usage, mobile app engagement, or Claude’s presence within third-party platforms. Anthropic’s actual reach is likely larger than the web traffic numbers suggest, particularly given its deep integration with AWS and growing presence in enterprise workflows.
Still, web traffic serves as a useful proxy for consumer mindshare, and on that metric, Claude’s March performance was exceptional.
The competitive dynamics are shifting in other ways too. DeepSeek’s decline from its January peak — when it briefly captivated the tech world with claims of training competitive models at a fraction of the typical cost — suggests that novelty alone doesn’t sustain market share. Users who flocked to DeepSeek out of curiosity appear to have migrated elsewhere, with Claude being one of the likely beneficiaries. The Chinese company faces additional headwinds in Western markets due to data privacy concerns and regulatory scrutiny that make enterprise adoption difficult.
Microsoft’s Copilot, despite having the enormous built-in distribution advantage of Windows, Office, and Edge, continues to underperform relative to its potential. Its market share decline to 3.6% is particularly notable given the billions Microsoft has invested in AI integration across its product line. The Copilot experience remains tightly bundled with Microsoft’s existing products, which may actually be limiting its appeal to users who want a standalone AI assistant rather than one embedded in a productivity suite they may or may not use.
Perplexity AI, which has carved out a niche as an AI-powered search alternative, held steady at around 3.4% market share in March. Its positioning is somewhat different from the pure chatbot competitors, focusing more on information retrieval with citations than on general-purpose conversation or code generation. But its stability while others declined suggests it has found a loyal user base.
What Comes Next
Anthropic isn’t resting on its March numbers. The company has signaled that Claude 4 is in development, with expectations of a release sometime in mid-2025. If the jump from Claude 3.5 to 3.7 was enough to catalyze a doubling of market share, a full generational upgrade could accelerate the trend further — assuming the model delivers meaningful improvements.
OpenAI, for its part, is preparing its own countermoves. The company is expected to release GPT-5 in 2025, a model that CEO Sam Altman has described as a significant leap forward. OpenAI also continues to expand ChatGPT’s feature set, recently adding more sophisticated image generation capabilities and deeper integration with external tools and data sources. The company’s installed base of more than 100 million weekly active users gives it a formidable moat, even as competitors chip away at the margins.
Google is pouring resources into Gemini’s next generation as well, with its DeepMind research division working on models that the company hopes will close the perceived quality gap with the best offerings from OpenAI and Anthropic. Google’s unique advantage — the ability to integrate AI directly into Search, the world’s most-used information tool — remains largely untapped in its full potential.
The broader market is also evolving in ways that could benefit challengers like Anthropic. Enterprise buyers are increasingly pursuing multi-model strategies, using different AI systems for different tasks rather than standardizing on a single provider. This approach reduces switching costs and makes it easier for a company like Anthropic to win specific workloads even within organizations that also use ChatGPT or Gemini for other purposes.
And then there’s the regulatory dimension. Anthropic has positioned itself as the safety-focused AI company, a brand identity that resonates with enterprise customers and government agencies that are wary of deploying AI systems without strong guardrails. As AI regulation tightens in Europe and gains momentum in the United States, that positioning could become a genuine competitive advantage rather than just a marketing message.
March 2025 may prove to be an inflection point. Not the moment Claude overtook ChatGPT — that remains a distant prospect — but the moment the market shifted from a near-monopoly to a genuine multi-player competition. Anthropic demonstrated that with the right product, at the right time, even a distant third-place competitor can rapidly close the gap.
The AI chatbot market is no longer a foregone conclusion. It’s a fight.
Claude’s Stunning Market Surge: How Anthropic Doubled Its Share and Shook the AI Chatbot Hierarchy first appeared on Web and IT News.
