Nicholas Domnisch is CEO / Managing Partner at EE Solutions, a technology operating partner for private capital firms and enterprises.
There are more private equity funds in the United States than McDonald’s restaurants.
KKR partner Alisa Wood made the comparison last year when she noted that there were “19,000 private equity funds in the US” versus roughly 14,000 McDonald’s restaurants. The point isn’t that there are 19,000 private equity firms. One sponsor can manage multiple funds across different strategies and vintages. The comparison shows the scale of capital and investment activity inside the industry.
That scale matters for AI because private equity also sits behind thousands of operating businesses. The American Investment Council says more than 13,000 U.S. companies are currently backed by private equity. Few industries have the same ability to test new operating capabilities across such a broad base of businesses and carry forward what works.
Much of the AI conversation so far has centered on productivity: faster diligence, automated reporting, lower support costs and more productive teams. All of that creates real value, but productivity may only be the first-order effect.
Wei Li, BlackRock’s global chief investment strategist, described AI in three phases: “The first phase is build-out, second phase is adoption and the third phase is broader productivity gain.” Her view is that the economy is still early in that progression, with evidence of gains in individual use cases but the broader effects still ahead, as she explained in an S&P Global interview.
For private capital, the important question is what that third phase actually looks like.
The economics of what is possible are changing.
McKinsey found that nearly 90% of organizations use AI in at least one business function, while only 7% report having fully scaled it. Eighty percent of respondents said efficiency was an objective of their AI initiatives, but companies seeing the most value were more likely to also pursue growth or innovation.
That distinction matters because companies make strategic decisions around constraints.
A customer segment can be too expensive to serve. A new product can require too much development. Personalization can be uneconomic outside the highest-value customers. An adjacent market can be too small to justify the people and systems required to enter it.
Those decisions may have been rational when they were made, but the underlying economics are changing.
Andreessen Horowitz general partner Anish Acharya has argued that cheaper software development has implications far beyond engineering productivity. In “Notes on AI Apps in 2026,” he argues that lower development costs have not yet fully changed how companies think about what should be built. In “Disposable Software,” he makes the related point that software which previously failed the economic test can now become viable.
When the cost of execution changes, the boundary of what is economically worth attempting changes with it.
Some old “nos” deserve another look.
A useful exercise for portfolio companies is to revisit initiatives rejected over the last several years because they were too expensive, labor-intensive, slow or complex. Most will likely still be bad ideas, but some will not be.
A service business may be able to support a larger customer base without scaling headcount at the same rate. A product that once required significant upfront investment may become cheap enough to test before committing major capital. Personalization that only made sense for the largest customers may become economical further down the market.
This is possibility expansion. Productivity improves the economics of what a business already does. Possibility expansion changes the set of things the business can economically do in the first place. For investors, that belongs in both value creation and underwriting.
The same shift helps competitors.
The reduction in execution costs doesn’t only benefit the portfolio company. A category that historically required substantial labor, expertise or operating complexity to enter may become easier for competitors to attack. A market that suddenly becomes attractive to one company can become attractive to several.
Competitive advantages are built within an economic environment, and AI is changing parts of that environment. If an investment case depends on scarce specialist capacity, expensive customization, high delivery costs or the difficulty of launching a competing offering, those assumptions deserve another look.
The upside case should be re-run. So should the downside.
Private capital can compound this across the portfolio.
Re-running these assumptions one company at a time is useful. The larger advantage comes from doing it across an entire portfolio.
A sponsor can observe where falling execution costs opened a new market in one company, changed a service model in another or created a competitive threat somewhere else. Those lessons don’t have to remain isolated.
Historically, operating knowledge has moved slowly. A lesson learned inside one company often stays there unless someone deliberately carries it elsewhere. AI can make that knowledge easier to capture and reuse because lessons that once lived across scattered documents, dashboards and teams can increasingly be summarized, searched and applied across the portfolio.
Over time, that can turn the portfolio into a learning system. One company tests an idea. Another improves it. A third applies the lesson in a different market. The sponsor gets better at recognizing which old constraints are disappearing and which new risks are emerging.
The opportunity isn’t simply to use AI in more companies. It’s to learn faster across more companies.
If the next phase of AI is broader productivity, private capital shouldn’t define productivity as doing the same work faster. The more important question is where changing economics expand what a company can attempt, where they weaken old assumptions and how quickly those lessons can compound across the portfolio.
The firms that build the fastest learning loop around those changes are the ones that I believe will be best positioned to capture the most value from them.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?