TL;DR
Vertical AI solutions are outpacing general-purpose AI in enterprise adoption because they offer domain-specific expertise, higher accuracy, and tangible ROI. Companies like Tempus in healthcare and Tractable in insurance demonstrate the power of focused AI applications. While vertical AI faces challenges like high development costs and limited scalability, its ability to deliver precise, actionable insights gives it a competitive edge. The next phase will see more hybrid models combining vertical and horizontal AI, but vertical dominance is here to stay.
The setup
Artificial intelligence is reshaping industries, but not all AI is created equal. For years, the enterprise world has debated the merits of vertical versus horizontal AI. Horizontal AI, or general-purpose AI, aims to solve a wide range of problems across multiple industries. Think of platforms like IBM Watson, which promises to tackle everything from customer service to medical diagnostics.
In contrast, vertical AI is tailored for specific industries or even particular use cases within those industries. It leverages deep domain knowledge to deliver highly specialized solutions. This approach is gaining traction, with companies like Tempus in healthcare, Tractable in insurance, and Uptake in industrial IoT leading the charge.
The conventional wisdom has long favored horizontal AI, driven by the allure of one-size-fits-all solutions. But the tide is turning. As enterprises demand more precise, actionable insights, vertical AI is emerging as the superior choice for many applications.
The argument
1. Domain-Specific Expertise Drives Higher Accuracy
Vertical AI's primary advantage lies in its domain-specific expertise. By focusing on a single industry or use case, these solutions can incorporate deep knowledge and context that general-purpose AI simply cannot match.
Example: Tempus, a leader in AI-driven cancer care, exemplifies this approach. The company has amassed one of the world's largest libraries of clinical and molecular data. Its AI algorithms are trained on this vast, specialized dataset, allowing them to deliver highly accurate insights for oncologists. In contrast, a horizontal AI platform would struggle to achieve the same level of accuracy without access to such specialized data.
Numbers: Tempus claims its AI can analyze genomic data at a rate of over 20,000 samples per week, with accuracy rates exceeding 90% in some diagnostic categories. This level of precision is critical in healthcare, where lives are at stake.
2. Tangible ROI and Faster Time to Value
Vertical AI solutions often deliver a clearer, faster return on investment. Because they are designed for specific use cases, they can be implemented more quickly and start generating value sooner.
Example: Tractable, an AI company focused on the insurance industry, offers a platform that automates the assessment of vehicle damage. Insurers using Tractable's solution have reported significant cost savings and efficiency gains. For instance, Tokio Marine, one of Japan's largest insurers, reduced its claim processing time by 50% after adopting Tractable's AI.
Numbers: Tractable claims its AI can process claims up to 10 times faster than traditional methods, with accuracy rates of over 95%. This kind of efficiency is a game-changer for insurers, who can now handle more claims with fewer resources.
3. Easier Integration and Customization
Vertical AI solutions are typically easier to integrate into existing enterprise systems. Because they are designed with specific industries in mind, they can more readily fit into established workflows and processes.
Example: Uptake, a company specializing in industrial AI, offers solutions that are tailored to the needs of manufacturers, energy companies, and other heavy industries. Its platform can be customized to fit the unique requirements of each client, making it easier to deploy and scale.
Counter-argument: Some argue that vertical AI's focus on specific industries limits its scalability. While it's true that these solutions may not be as universally applicable as horizontal AI, their ability to deliver precise, actionable insights often outweighs this limitation. Moreover, many vertical AI companies are exploring ways to expand their offerings while maintaining their domain-specific expertise.
What the conventional view gets wrong
The conventional view often underestimates the complexity of enterprise AI needs. It assumes that a single, general-purpose AI platform can meet the diverse requirements of different industries. This assumption is flawed for several reasons.
1. Overlooking the Importance of Domain Knowledge
General-purpose AI platforms lack the deep domain knowledge that is crucial for many enterprise applications. For example, in healthcare, understanding the nuances of medical terminology, patient history, and treatment protocols is essential for accurate diagnostics and treatment recommendations. A horizontal AI platform simply cannot replicate this level of expertise without extensive customization.
2. Ignoring the Variability of Enterprise Data
Enterprise data is highly variable, with different industries and even different companies within the same industry having unique data structures and requirements. A horizontal AI platform may struggle to handle this variability, leading to suboptimal performance and inaccurate results.
3. Underestimating the Need for Customization
Enterprises often require highly customized solutions to meet their specific needs. A one-size-fits-all approach may work for some applications, but for many others, it falls short. Vertical AI solutions, with their focus on specific industries and use cases, are better equipped to deliver the customization that enterprises demand.
The implications
For Builders
Builders of AI solutions need to recognize the growing demand for vertical AI. This means investing in domain-specific expertise and data collection. It also means developing flexible platforms that can be easily customized for different clients and use cases.
For Users
Enterprises should carefully evaluate their AI needs and consider whether a vertical AI solution might offer better performance and ROI than a general-purpose platform. They should also be prepared to invest in the necessary data and infrastructure to support these solutions.
For Regulators
As vertical AI becomes more prevalent, regulators will need to develop frameworks that address the unique challenges and opportunities of these solutions. This includes ensuring data privacy, addressing bias and discrimination, and promoting transparency and accountability.
For the Industry
The rise of vertical AI could lead to a more fragmented AI landscape, with different solutions catering to different industries and use cases. This could create opportunities for collaboration and integration, as companies seek to combine the strengths of vertical and horizontal AI.
What I'm watching next
1. Hybrid Models Combining Vertical and Horizontal AI
I expect to see more companies exploring hybrid models that combine the strengths of both vertical and horizontal AI. For example, a healthcare AI platform might use a horizontal AI engine for basic diagnostics and a vertical AI component for more specialized tasks like treatment recommendations.
2. Expansion of Vertical AI into New Industries
As vertical AI proves its value, I anticipate its expansion into new industries. Sectors like agriculture, finance, and education are likely candidates, as they have unique challenges that could benefit from specialized AI solutions.
3. Increased Investment in Domain-Specific Data
The success of vertical AI hinges on access to high-quality, domain-specific data. I expect to see more investment in data collection and curation, as companies seek to build the datasets needed to train and refine their AI models.
The bottom line
Vertical AI is winning the enterprise space because it offers the precision, accuracy, and ROI that enterprises demand. While it faces challenges like scalability and high development costs, its ability to deliver tangible benefits makes it a compelling choice for many applications. As the AI landscape evolves, we can expect to see more hybrid models that combine the strengths of both vertical and horizontal AI. But for now, the dominance of vertical AI is a trend that is here to stay.
In the coming years, the winners in the AI space will be those who can effectively balance the broad applicability of horizontal AI with the specialized expertise of vertical AI. Companies that invest in domain-specific knowledge, high-quality data, and flexible platforms will be well-positioned to thrive in this new era. The future of AI is vertical, and the enterprise world is taking notice.
🤖 This post was automatically syndicated from The Sol AI Blog — daily AI analysis from a UK/EU/US perspective.
Follow along for more →
