A few years into the generative AI explosion, we’ve reached a critical inflection point. The public is intimately aware of AI and its power to transform nearly every aspect of work and life, once revolutionary apps like ChatGPT and Copilot have become almost quotidian, and glimmers of true enterprise transformation are appearing. Humans and AI are dancing together.
But this dance is more akin to a line dance than a waltz, especially as this wave of AI enters the enterprise. Too many AI companies are pursuing generalized applications, with little understanding or integration into existing workflows, and, problematically, selling point solutions that clutter enterprise buyers’ minds in a manner reminiscent of the SaaS era, where 20 point solutions didn’t “talk” to each other. We owe it to ourselves not to repeat that mistake.
If enterprises want to avoid the 95% failure rate shown by the now-famous MIT report, they need to do 3 critical things:
1) Partner with an AI company that deeply understands their workflows, as a lack of proper integration hinders success.
2) Target applications that revolve around their most valuable “transactions” and the corresponding gold mine of data sets. This enables the first step toward an AI ‘operating system.’
3) Partner with an AI leader who has a connected AI vision of the future of their enterprise and industry, because everything is changing rapidly.
This last point deserves further explanation. If AI is to be maximally impactful, it must sense like a human. In particular, I’m focused on helping AI SAY (Voice AI), SEE (Vision AI), and DO (Robotics), then weave these senses together in a finely coordinated fashion—much like humans do. This is the connected AI vision that every industry must adopt: SAY-SEE-DO.
It is very clear AI is creating tangible and lasting value for Enterprises in every industry, but in order to unlock maximal value and set your company up for success in a future-state AI-native world, it’s critical to ditch the SaaS vendors and embrace the AI visionaries who are building specifically for your industry.
Enterprise data: the vertical AI gold mine
To launch end-to-end automation within a given industry, you must first examine the “gold mine” of data in each enterprise. In restaurants, for example, the gold mine of data is the end customer. What are diners doing? How are they interacting with the enterprise? In this industry, the end customer is speaking with restaurant staff – to place their orders, ask questions about the menu, and provide feedback in real-time. The automation entry point for restaurants, thus, is Voice AI.
The data gold mine looks a little different in other industries. Take, for example, agribusinesses. In this industry, the most prized data revolves around individual plants: are they diseased? Are they water stressed? Are they damaged in some way? Access to this dataset empowers various constituencies: it allows insurers to insure more effectively, lenders to lend more confidently, etc. The most meaningful AI application for agribusinesses, then, is not Voice AI, but Vision: a tool that allows growers and interested parties to see, at an individual plant level, what’s going on in the soil. In the construction industry, the data gold mine revolves around machinery: its abilities, its useful life, and its usage. In this industry, the most meaningful entry AI application is likely robotics.
When we think about data as the backbone of the body, we can then determine which AI application – Voice, Vision, or Robotics – will allow a given enterprise to leverage the most data. That is the first step toward building the automation operating system.
Our Investment Edge
America’s industries are adopting AI at a rapid pace. A huge portion of enterprise spend is currently tied up in places that only AI can touch. It’s critical, thus, for each enterprise to select the AI entry point that will best leverage its data goldmine and set it squarely on the path to streamlined, connected automation.
The next 24 months represent a critical juncture for companies carving out leadership roles in each of these categories. In 2025-2026, businesses that attempt to rely on a thin horizontal wrapper to solve vertical-specific problems without deep data and integration expertise will fail. Conversely, companies that combine deep domain data, robust operational frameworks, and enterprise integrations will pull ahead. As enterprises realize they cannot build competitive systems fast enough in-house, they will seek to acquire proven vertical players. That’s where our focus and expertise lies.
Voice AI is on the cutting edge of this vertical AI revolution—and our experience in this sector can serve as a replicable paradigm that will inform further scale in voice and growth in vision and robotics. Our history with Voice AI spans 15 years, most relevantly in scaling Presto’s Voice AI business: thanks to our work at Presto, we understand the nuances of conversational flow, human-in-the-loop, enterprise integration, customer success/ROI, and product roadmap pitfalls better than many of the firms that are now crowding this space. This allows us to understand more immediately which vertical voice companies are set to win.
In certain other industries, Vision and Robotics may be the more first-order application of AI. For example, in agriculture, the plant-level data is the gold mine and in construction, the physical data is the gold mine—so Vision and Robotics, respectively, are the foundational AI layer.
It’s our plan to back those businesses early and own the picks and shovels of the Vertical AI revolution.
The Opportunity Ahead
Vertical AI is the next great platform shift in technology. The last decade was defined by horizontal cloud and SaaS; the next will be defined by vertical AI stacks that can SAY, SEE, and DO as well as—and in some cases better than—human workers.
As we approach the moment in the cycle when vertical AI moves from pilot to production, we believe the firms and founders who can truly execute amid the messy realities of specific industries can define the next generation of companies that are built to last.