Introduction: A recent report from Menlo Ventures heralds Anthropic’s supposed dethroning of OpenAI in enterprise AI usage, signaling a dramatic shift in the highly competitive LLM landscape. But before we declare a new monarch in the AI realm, it’s crucial to scrutinize the data’s foundations and the inherent biases in such early-stage market analyses.

Key Points

• Anthropic is reported to have surpassed OpenAI in enterprise LLM market share by usage (32% vs. 25%), with a particularly strong lead in coding applications.
• This represents a rapid and significant reversal from just two years ago, underscoring the extreme volatility and rapid evolution of the enterprise AI sector.
• A critical aspect of Anthropic’s reported surge hinges on the performance and adoption of models, some of which were only released recently (June 2024) or are still in the future (February 2025), raising questions about the real-time validity of “current usage” metrics.

In-Depth Analysis

The narrative spun by the Menlo Ventures report paints a picture of swift corporate preference shifting towards Anthropic, particularly for its Claude models. For veteran observers of the technology sector, such abrupt market reversals, while not unprecedented, demand a closer look beyond the headline numbers. Two years ago, OpenAI commanded a whopping 50% of the enterprise market, with Anthropic trailing at 12%. Today, we’re told, those figures are neatly inverted. This dramatic swing, if truly reflective of deeply embedded enterprise adoption, would imply a significant re-evaluation of core LLM capabilities and vendor relationships.

Why might enterprises be shifting? The report vaguely attributes Anthropic’s surge to the release of Claude 3.5 Sonnet (June 2024) and the future Claude 3.7 Sonnet (February 2025). This is where skepticism must kick in. How can a model released just last month, and another not yet available, account for a substantial portion of current market share by usage? This suggests the figures might be heavily weighted by pilot projects, API calls in early development, or perhaps even a significant dose of projected uptake rather than fully scaled, production-level deployments. Menlo Ventures, as a venture capital firm, has an inherent interest in identifying and championing market leaders, which can sometimes lead to reports that lean optimistic. “Usage” itself is a broad term; does it signify extensive integration across business units, or merely experimentation by development teams?

Anthropic has certainly cultivated a reputation for safety, guardrails, and longer context windows, attributes that resonate with risk-averse enterprise clients. Their focus on constitutional AI and responsible development could be a genuine differentiator for companies navigating complex regulatory and ethical landscapes. The specific mention of Anthropic’s dominance in coding (42% vs. OpenAI’s 21%) suggests that developers might prefer Claude’s output for code generation, completion, and debugging—a high-value, high-volume use case for many businesses. Furthermore, the report’s finding that enterprises overwhelmingly prefer closed-source models over open-source alternatives (with only 13% daily workload usage for the latter) speaks to the perceived reliability, support, and perhaps easier integration path offered by vendors like Anthropic and OpenAI compared to managing open-source deployments internally. However, true market dominance in the enterprise isn’t just about features; it’s about robust support, deep integrations, competitive pricing, and a proven track record of scaling. These are factors not easily captured by simple “usage” percentages.

Contrasting Viewpoint

While the report touts Anthropic’s enterprise gains, a counter-narrative quickly emerges when considering the broader AI ecosystem. OpenAI, despite its reported enterprise decline, remains a formidable force with a staggering consumer footprint, reporting 2.5 billion ChatGPT prompts daily. This massive consumer interaction not only provides an invaluable data flywheel for model improvement but also signals a powerful brand presence and public mindshare that Anthropic simply doesn’t command. Enterprise market share, especially in such a nascent and rapidly evolving field, can be notoriously fluid. Pilot projects don’t always translate into full-scale deployments, and switching costs for LLM APIs are not yet prohibitively high. Furthermore, while Anthropic emphasizes safety, OpenAI has the immense backing of Microsoft, leveraging Azure’s vast enterprise infrastructure and sales channels—a strategic advantage that cannot be understated. One must also consider the methodology: Is “usage” measured by token count, number of API calls, or active user licenses? Without transparent methodology, it’s difficult to ascertain if Anthropic’s lead is truly a reflection of deeper, embedded enterprise adoption, or merely a successful push in specific high-profile early adopters.

Future Outlook

The next 12-24 months in enterprise AI will undoubtedly see continued intense competition, with the reported Anthropic surge serving as a potent reminder of the market’s dynamism. The biggest hurdles remain less about raw model performance and more about operationalizing AI within complex, often legacy-laden enterprise environments. Issues of data governance, security, explainability, and the elusive “last mile” integration into business workflows will dictate long-term success far more than benchmark scores. We’ll likely see increased differentiation beyond general-purpose models, with vendors specializing in vertical-specific applications or offering highly customizable solutions. The role of open source, despite the current reported decline in enterprise adoption, is far from over; as models mature and more companies develop in-house expertise, the cost-effectiveness and control offered by open-source solutions will undoubtedly resurface as a compelling value proposition. Ultimately, the true victor in the enterprise AI race will be the provider that can not only deliver cutting-edge models but also build the most robust, secure, and easily integrated platforms that demonstrably drive business value at scale.

For more context on the operational challenges of deploying AI in complex organizations, see our deep dive on [[The Pitfalls of Enterprise AI Adoption]].

Further Reading

---

Original Source: Enterprises prefer Anthropic’s AI models over anyone else’s, including OpenAI’s (TechCrunch AI)

阅读中文版 (Read Chinese Version)