A new a16z report looks at which AI companies startups are actually paying for | TechCrunch
Key Takeaways
- a16z released its first AI Spending Report using Mercury transaction data to analyze startup spending on AI applications.
- Startups are currently adopting a wide proliferation of different AI tools rather than consolidating around a few leaders in each category.
- Spending is heavily focused on 'human augmentors' or 'copilots' to boost productivity, indicating a slow shift toward full end-to-end agentic workflows.
- OpenAI and Anthropic dominate the top spending slots, and consumer tools like CapCut and Midjourney are being adopted by enterprises.
- The market shows room for many different companies, with horizontal applications making up at least 60% of the spending list.
Andreessen Horowitz (a16z) released its first AI Spending Report, utilizing transaction data from fintech firm Mercury to analyze the top 50 AI-native application layer companies where startups are spending money. Partners Olivia Moore and Seema Amble noted that the data reflects a current proliferation of diverse AI products, indicating that the market has not yet consolidated around one or two dominant tools per category. A significant portion of spending is directed towards 'human augmentors' or 'copilots' intended to enhance workforce productivity, implying that startups are hesitant to fully transition to end-to-end agentic workflows. OpenAI secured the top spending spot, followed by Anthropic, while coding tools like Replit were also highly ranked, though consumer-facing apps like CapCut and Midjourney are surprisingly being adopted by enterprises. The report suggests that while horizontal applications dominate spending, AI is making inroads into verticals previously difficult for software to crack, and the trend is expected to shift from copilots toward full agent tools as technology improves.
