Leading technologists and CEOs now openly forecast a future where AI drives a productivity explosion large enough to make work optional and fund something they call universal high income. The debate around those forecasts is loud, personality-driven, and mostly unproductive. Strip out the personalities and a harder question is sitting underneath, one that almost nobody in the debate is answering: if AI really does create an enormous economic surplus, whose balance sheet does it land on?
That question has a practical answer, and it does not require you to believe any particular forecast. It requires you to look at where the surplus is already showing up, who is already capturing it, and what position you would need to hold to be on the receiving end. This piece lays out the framework we use at REV Global: the difference between waiting for universal high income and building what we call private high income through AI-augmented, tax-advantaged ownership of mid-market businesses.
The Forecast, and Why It Is Suddenly Serious
Grand claims about AI are not new. What changed over the past eighteen months is that the most conservative institutions in the economy started acting on them. As we covered in our analysis of bank workforce cuts, major Wall Street firms are reducing graduate analyst hiring by as much as two-thirds as AI absorbs entry-level work. A recent Oliver Wyman survey found more than 40% of global CEOs plan to cut junior roles within two years. McKinsey's displacement research estimates roughly 30% of hours worked in the US economy could be automated by 2030.
You do not have to like the forecasters, and you do not have to accept the utopian version of the story. The signal is not in the keynotes. It is in the hiring plans, the org charts, and the unit economics that the least sentimental institutions in the world have already re-run. Regardless of how one views any particular messenger, the underlying economic shift is clear: the cost of a defined and growing category of knowledge work is collapsing, and the value of that work is being transferred somewhere.
The Question Nobody Is Answering
Here is the part the public debate keeps skipping. Productivity gains are not abstract. When AI does in minutes what a salaried employee did in days, the difference between the old cost and the new cost becomes surplus, and surplus always lands on a specific balance sheet. It accrues to whoever owns the system where the work happens: the business owner whose margins expand, the acquirer who underwrote the improvement, the shareholder of the platform that sold the automation.
A W-2 paycheck sits on one side of that ledger. Ownership sits on the other. That is not a moral statement, and it is not a prediction about whether any government will eventually redistribute some of the surplus as universal income. It is an accounting identity. In the years between now and whatever policy future arrives, the surplus flows to owners by default.
Universal High Income vs. Private High Income
Universal high income, if it ever arrives, will be a policy outcome. It depends on legislation, on political consensus, on a distribution mechanism that does not exist yet, and on a timeline you do not control. It might happen. Betting your family's financial position on it is a different matter.
Private high income is the same underlying idea captured through a different mechanism: instead of waiting for the AI surplus to be redistributed to you, you own the assets where the surplus is generated. In practice, for the investors and operators we work with, that means owning cash-flowing mid-market businesses where AI measurably expands margins, structured so that the after-tax outcome is as strong as the pre-tax one. The thesis reduces to a simple equation: ownership plus AI plus tax advantage equals private high income.
The Three Execution Layers
The equation only works if all three layers are executed deliberately. Here is how each one contributes.
Layer one: ownership of the right businesses. Not every business benefits from automation. The targets that do best in our experience are essential services in the lower middle market: HVAC, electrical, plumbing, commercial cleaning, logistics. Demand for these services does not switch off when AI arrives; a broken compressor still needs a technician. What AI changes is the cost structure around the technician: dispatch, scheduling, follow-up, invoicing, AR. We published a full target list in Top 10 Businesses to Buy and Invest In as a W-2 Employee in the AI Era.
Layer two: a Day-1 AI stack. The businesses we acquire and advise deploy large language models, autonomous agents, and proprietary models from the first week of ownership, not as a year-three digital transformation project. The mechanics are detailed in our 100-Day AI Integration Blueprint. In our engagements, that compression is the difference between waiting eighteen months for value creation and seeing it inside a single quarter:
- Time to value: roughly 90 days to capture what historically took 18 months of operational improvement.
- Growth multiples: AI-augmented rollups in our models run at approximately 3.2 to 3.3x the growth trajectory of their un-augmented baselines.
- Exit premium: AI-native mid-market businesses are commanding premium multiples from buyers who no longer want to fund the transformation themselves.
Those figures are illustrative ranges based on REV Global engagements and deal models; results vary by business, market, and execution quality.
Layer three: tax-advantaged structure. Capturing the surplus is only half the job; keeping it is the other half. The current tax code offers ownership structures that materially improve after-tax outcomes for qualifying acquisitions and investments, and structuring the deal correctly from the start is often worth as much as the operational improvement itself. This is planning-intensive work that depends on individual circumstances, which is exactly why it belongs in the underwriting conversation rather than as an afterthought at exit.
The Skeptic's Version Still Works
A fair number of the owners and acquirers we talk to are AI cynics, and we consider that healthy. So run the skeptic's math. Assume the grand forecasts are wrong. Assume no productivity explosion, no optional work, no universal anything. Assume AI turns out to be nothing more than a set of modestly useful tools that shave overhead in ordinary businesses.
Under that deflated assumption, the thesis still clears. A boring essential-services business that uses modest AI tooling to cut back-office cost and lift technician utilization is a stronger business than the one next door that did not. Margins expand somewhat instead of dramatically. The exit multiple improves somewhat instead of dramatically. You still own a better company than you bought. If the skeptics are right, you own a stronger business; if the optimists are right, you own the upside. There is no version of the forecast where owning the productive asset is the losing position, which is precisely what makes ownership the hedge.
What This Means for Owners and Acquirers
For the executive or professional watching entry-level work get absorbed, the takeaway is not panic; it is position. The window where under-digitized mid-market businesses can be acquired on trailing financials, before AI-driven margin expansion is priced into every deal, is open now and will not stay open. Sophisticated buyers are already underwriting the improvement at the LOI stage.
For the owner already operating a services business, the same logic runs in reverse: every quarter you run un-augmented, you are accumulating a gap against the acquirer or competitor who is not. The good news is that closing the gap no longer requires a custom build or an IT department. It is a configuration exercise with a measurable payback period.
The world will keep debating universal high income, and the debate will keep generating more heat than light. The practical move is quieter: own the assets where the surplus lands, augment them from day one, structure them so you keep what they earn. That is private high income, and it does not require anyone's forecast to come true.