Vertical AI: why industry-specific software is attracting founders
As foundation models become widely available, many founders are building AI for specific industries rather than general tools. Here is why vertical AI is attracting founders and investors, and where the risks lie.
Key points
- Crunchbase reported AI took close to half of global venture funding in 2025, and over 42% of seed funding.
- Vertical AI applies models to specific industries, using domain expertise and proprietary data.
- Adoption is broad but shallow; companies that help customers scale AI may have an advantage.
- Risks include smaller markets, dependence on model providers and competition from incumbents.
The largest sums in AI have gone to companies building foundation models, the general-purpose systems that underpin chatbots and other tools. But a growing number of founders are taking a different path. They are building AI products for specific industries, such as legal services, health care, construction, agriculture, mining and financial services. This approach is often called vertical AI.
The scale of AI investment
Artificial intelligence has become the dominant theme in venture capital. Crunchbase reported that AI companies raised about US$202 billion in 2025, close to half of all global venture funding. Foundation model companies took about 40% of that. At the seed stage, Crunchbase found that more than 42% of global seed funding in 2025 went to companies in AI-focused categories, up from 30% in 2024.
In Australia, the State of Australian Startup Funding report by Cut Through Venture and Folklore Ventures found that companies building with AI in their technology stack attracted 61% of total funding in 2025, according to Capital Brief.
Why vertical AI appeals
Vertical AI companies argue that general-purpose models are becoming widely available, so lasting value will come from applying them deeply to specific problems. Their potential advantages include:
- Domain expertise. Founders who know an industry can design products that fit real workflows, regulations and terminology.
- Proprietary data. Industry-specific data, such as claims records, maintenance logs or clinical notes, can improve products in ways competitors cannot easily copy.
- Clear buyers and budgets. Selling to a defined industry can make it easier to identify decision makers and quantify value.
- Workflow integration. Products embedded in daily operations can be harder to replace than standalone tools.
Adoption is the challenge
Enterprises are adopting AI broadly but unevenly. McKinsey’s State of AI 2025 survey found that 88% of organisations use AI in at least one business function, but only about one-third have begun to scale it across the enterprise. Vertical AI companies that help customers move from pilots to production, with integration, training and measurement, may be well placed.
The risks
Vertical AI is not without risk:
- Smaller markets. A single industry may limit the size of the opportunity, particularly in a small domestic market.
- Model dependence. Many vertical AI products rely on third-party models, which affects costs and differentiation.
- Incumbent response. Established software vendors in each industry are adding AI features to their own products.
- Regulation. Industries such as health and finance have strict rules on data, privacy and accountability.
Regional opportunities
Australia’s economy offers natural opportunities for vertical AI in resources, agriculture, energy, health and financial services, industries where local expertise is deep and the problems are global. In the United States, the size of each vertical market can support large companies. In Europe, regulatory frameworks such as the EU AI Act shape how vertical AI products are built and sold. In Asia, manufacturing, logistics and financial services offer large markets for industry-specific AI.
What investors look for
Investors assessing vertical AI typically ask how the product changes a workflow, what proprietary data the company accumulates, how margins are affected by model costs and whether the company can expand into adjacent verticals or geographies.
Frequently asked questions
What is the difference between vertical and horizontal AI?
Horizontal AI serves many industries with general tools, such as writing assistants. Vertical AI targets a specific industry with tailored workflows and data.
Is vertical AI defensible?
It can be, if the company builds proprietary data, deep workflow integration and strong customer relationships. Defensibility based only on access to a model is weak.
Can Australian vertical AI companies compete globally?
Yes, particularly in industries where Australia has deep expertise, such as mining, agriculture and energy, though they need to plan for international expansion early.
References
- Crunchbase News: 6 charts that show the big AI funding trends of 2025
- Crunchbase News: Seed funding in 2025 broke records around big rounds and AI
- Capital Brief: Aussie startups in 2025 had strongest funding quarter in years
- McKinsey: The state of AI in 2025
Related reading
General information only. This article reflects the author’s views at the time of writing. It does not take into account your objectives, financial situation or needs, and it is not financial, legal or tax advice or a recommendation to invest. Figures are from the sources named and may change. See our terms of use.