AI’s mid-market moment: why the next growth revolution will come from the middle

For decades, mid-market businesses – the “missing middle” between startups and multinational corporations – have been a crucial growth engine for the global economy. They account for roughly one-third of private-sector GDP and employment in developed economies like the US and make a significant contribution in emerging markets.

Perhaps most importantly, they have been resilient through economic downturns: from 2007 to 2010, for example, US mid-market businesses created 2.2 million jobs, while large businesses shed 3.8 million jobs. Again, in the US, from the beginning of 2020 through the end of 2021, these businesses grew their employee base by nearly 10%, while large businesses shrank by nearly 1%.

Despite their strengths and dynamism, mid-market businesses often have weak IT infrastructure and systems compared to larger competitors. In fact, only one in seven US mid-market businesses on average would historically allocate an incremental investment dollar to technology, compared to other uses.

There are three reasons why I believe AI is about to unlock the potential of the mid-market even further:

In combination, this means smaller firms can potentially access world-class knowledge, consulting frameworks and analytics once reserved for large corporations. We see this in our portfolio at Brightstar Capital Partners, where one of our businesses has created a proprietary module that can turn 30 terabytes of data into tagged and retrievable datasets within 30 minutes. In this new “AI power play”, we see competitiveness being democratized – enabling a more inclusive and dynamic economy.

There is a lot of value at stake: a 2023 study estimates the global economic potential of Gen AI to be between $6 trillion and $8 trillion. With mid-market companies accounting for a third of private-sector GDP – and historically higher growth rates than large companies – it is a fair assumption that they can capture at least $2 trillion of that value. This is equivalent to the size of Canada’s GDP and doesn’t even include use cases beyond Gen AI, for example, in the more “traditional” domains of machine learning.

Capturing this value would amount to a tremendous infusion of growth into local economies, where the majority of mid-market businesses operate and maintain strong roots in the community. It would contribute to local employment and allow established businesses to offer new services and products – unlocking further new sources of growth for the economy.

With AI as a significant factor in transforming nearly a quarter of jobs by 2027, the challenge is no longer whether AI will change work, but how. For mid-market firms, opportunity lies in redeploying talent from routine tasks toward creative and innovative ones. I am confident that established businesses will benefit from AI, capitalizing on their resilient business models, proprietary data sets, financial resources and established brands, distribution and channels. But, with up to 95% of AI pilots failing to date, they need to overcome significant adoption challenges.

Private equity investors can play a role here: connecting companies to proven use cases, technology partnerships and implementation know-how that make responsible scaling possible. Their ability to provide patient capital and to closely partner with management teams allows for more effective implementation than would be possible in a public company.

At Brightstar, we are leaning into Gen AI, aiming to be early adopters in our industry. We have built a portal of more than 30 AI agents on the OpenAI platform to assist our staff in daily activities and decision-making. Very early on, we chose to augment rather than automate – partly because the systems still need human supervision, but more importantly because we believe the real advantage of AI will be to improve human decision-making and creativity.

As an example, we have deployed a “Red Team Agent” that can craft a contrarian case for every investment opportunity we consider – both to provide an objective, alternative viewpoint and to surface potential issues the deal team might have missed. The goal is to improve our decision-making pre-acquisition, ensure diversity of thought in our conversations and inform our plans for value creation post acquisition.

We believe the upside of growth is a more powerful lever than traditional automation. As such, in our car auction business we are focused on deploying AI to help marketing teams create more effective and targeted campaigns that connect customers to the inventory in which they are most interested. We give our portfolio companies access to the AI training materials we’ve used for our own staff, so they can improve AI literacy across their workforce.

The balance between workforce augmentation and automation will determine if AI expands opportunity or deepens inequality. If each company determines that balance in isolation, there is a danger that companies eliminate entry-level jobs for short-term benefit but to the long-term detriment of the talent pipeline. Finding the optimal approach requires genuine collaboration among businesses, investors and policy-makers in the spirit of the World Economic Forum's multistakeholder approach.

Just as industrial and trade apprenticeships created a resilient workforce in previous times, an AI-era equivalent supported by public and private actors can ensure this transformation benefits workers, firms and societies alike.