门罗风投合伙人蒂姆•塔利（Tim Tully）在接受采访时表示，这种权力转移在一定程度上要归功于Claude 3.5的进步，因为大多数公司都在使用三个或更多的大型人工智能模型。他说，尽管OpenAI和Anthropic主导了公司的人工智能模型使用，但人们正在“玩弄模型”，而这种习惯“并不是一个很好的理解数据”。
Data released by Menlo Ventures on Wednesday shows that spending by companies on generative ai has soared by 500%, rising from $2.3 billion in 2023 to $13.8 billion.
The report also found that OpenAI's market share in the enterprise ai sector dropped from 50% to 34%. Anthropic's market share doubled, increasing from 12% to 24%. The report stated that this result comes from a survey of 600 corporate IT decision-makers, with companies employing more than 50 people.
Menlo is an investor in Anthropic.
Menlo Ventures partner Tim Tully stated in an interview that this shift of power is partly attributed to the advancements of Claude 3.5, as most companies are using three or more large ai models. He mentioned that although OpenAI and Anthropic dominate the use of corporate ai models, people are "playing with models," and this habit is "not a good way to understand data."
Tully explained, "Developers are very savvy—they know how to switch quickly between models." "They choose the model that best fits their use case... most likely Claude 3.5."
Meta's market share remains at 16%, while Cohere's market share stays at 3%. Google's share increased from 7% to 12%, whereas Mistral's market share dropped by one percentage point, falling to 5% in 2024.
The report found that foundational models such as OpenAI's ChatGPT, Google's Gemini, and Anthropic's Claude still dominate enterprise spending, with large language models receiving $6.5 billion in corporate investment.
Monroe's report is bullish on ai agents, which represent the leading ai trend and investment area for 2024. Google, microsoft, amazon, OpenAI, and Anthropic are all pursuing this technology. ai agents are seen as a step beyond chatbots. They can perform multi-step, complex tasks on behalf of the user and generate their own to-do lists, so users do not have to guide them through the entire process step by step.
The report found that code generation is the primary use case for generative ai, with over half of respondents listing it as the main application. This is followed by support chatbots at 31%, and then by enterprise search and retrieval, data extraction and transformation, and meeting summaries.
