Enterprises prefer Anthropic’s AI models over anyone else’s, including OpenAI’s

AI research lab Anthropic has surpassed OpenAI in enterprise large language model (LLM) market share by usage, now holding 32% compared to OpenAI's 25%, marking a significant reversal from two years ago when OpenAI led with 50%. This shift, notably more pronounced in coding applications where Anthropic commands 42% to OpenAI's 21%, is attributed to the strong performance and adoption of Anthropic's recent Claude 3.5 and 3.7 Sonnet models. While OpenAI retains its dominant position in the consumer market, the Menlo Ventures report underscores enterprises' increasing preference for closed models, with open-source LLM adoption declining.

Analysis

A significant shift is underway in the enterprise AI market, with Anthropic capturing the leading position from OpenAI. According to a Menlo Ventures report, Anthropic now commands 32% of the enterprise large language model (LLM) market share by usage, compared to OpenAI's 25%. This marks a stark reversal from two years prior when OpenAI held a dominant 50% share to Anthropic's 12%. The catalyst for this rapid ascent appears to be strong enterprise adoption following the release of Anthropic's Claude 3.5 and 3.7 Sonnet models. The company's lead is even more pronounced in the high-value coding segment, where its 42% market share is double that of OpenAI's 21%. While OpenAI maintains a strong foothold in the consumer market, the report highlights a clear enterprise preference for closed models, with usage of open-source alternatives declining from 19% to 13% of daily workloads in 2025. This trend benefits Anthropic and OpenAI but poses a challenge for Meta, which remains the leader in the shrinking enterprise open-source segment. Google is also noted as a beneficiary of this trend, with its enterprise model usage increasing.

Key Decisions for Investors

• Investors should recognize the high competitive velocity in the AI sector, as the rapid shift in market share from OpenAI to Anthropic demonstrates that product release cycles can quickly disrupt established leaders.
• The data presents a positive signal for Alphabet's (GOOGL) enterprise AI strategy, which is reportedly gaining traction, but suggests a potential headwind for Meta Platforms (META) as enterprise preference shifts away from the open-source models where it dominates.
• It is prudent to re-evaluate investment theses that are heavily predicated on sustained market dominance by any single AI provider, particularly in the enterprise segment, which is proving to be more fragmented and competitive than the consumer market.