Nick Leopard, Founder & CEO, Accordion.
Venture capital moves fast and breaks things. Private equity moves deliberately and fixes them. That difference explains a lot about where PE sits on AI right now: not absent, but methodical in a moment moving faster than methodical allows.
According to my company’s research, VC-backed companies have adopted AI at a 77% rate. PE-backed companies are at 59%. That gap is less about technology than organizational readiness.
And inside that 59%, the picture gets worse: 98% of sponsors say AI is a priority, but fewer than 1 in 3 CFOs have acted on it in any way that shows up in results. The gap isn’t about access but about what happens after deployment, where PE keeps getting stuck.
I’ve seen this pattern in enough portfolio companies to recognize it immediately: Variance commentary gets generated in seconds and then rebuilt in Excel. Forecasts refresh daily and get reviewed once a week. Pricing recommendations arrive in real time and wait for approval thresholds set three quarters ago.
This is the adoption gap. Every quarter it persists is a quarter of wasted value creation.
The limiting factor isn’t the tech but the organization around the tools: fragmented data, undocumented processes, parallel workflows that quietly survive every new tool deployment because no one pulled the old one out.
The cost of AI is up front, but its value is downstream. A recent Bain & Company survey of CFOs found that only 31% rate AI outcomes in finance as strongly positive, even as investment accelerates. Accenture’s analysis of PE portfolio companies puts the upside in sharp relief: Every $1 invested in AI transformation can deliver an annualized EBITDA uplift of 2x to 4x at exit. But that return is conditional: It goes to companies that close the organizational gap first.
Most portfolio companies aren’t there yet. According to Google Cloud’s 2025 ROI of AI in Manufacturing report, AI gains are concentrated among companies that have built cross-functional collaboration into the deployment from the start.
Today’s AI systems can help plan multistep tasks, navigate enterprise software, synthesize information across sources and surface decisions before humans would know to ask. Capability isn’t the problem; absorption is.
Closing The Adoption Gap: How Finance Changes Shape
The shape of finance changes when the function stops producing answers on a schedule and starts moving with the business. When demand shifts, outlooks update instantly, forcing decisions on pricing or capacity while there’s still time to act. Cash positions stop being reported at period end and start driving decisions as conditions change.
That’s a fundamentally different way of working, where the team’s job shifts from assembling the answer to deciding what to do with it. What we’re talking about is building AI-enabled operating systems around the finance function: ones designed from the start for adoption and measurable ROI, embedded directly into the workflows where value is created or lost.
Five Things The Teams That Get There Do Differently
Those that close the gap share a few things in common. They:
1. Lock a small number of metrics first.
AI produces outputs that come back quickly but cannot be used without a fact check. Teams trace the numbers, rebuild parts of the analysis and only then move forward. The fix is to take one metric and make it reconcile end-to-end before touching anything else. Once an output can be trusted as-is, behavior changes.
2. Pull decisions forward.
Most AI efforts improve individual steps without changing how work actually moves. Reports generate faster, but processes around them stay the same. The shift comes when teams identify where decisions are actually made and restructure the work around that point: removing handoffs rather than accelerating them, using forecasts continuously rather than refreshing them.
3. Tie use cases to financial levers before scaling.
Forecast precision that doesn’t change a cash decision is not a use case; it is a feature. In a PE environment, the connection to EBITDA must be explicit up front. If it cannot be drawn clearly, the use case does not move forward.
4. Eliminate old processes rather than running them in parallel.
Adoption stalls when AI becomes a reference point rather than the decision layer. The fallback has to go. At some point, teams stop checking the output and start using it, even when it’s not perfect. Until that happens, the old process is still the real process.
5. Make outputs defensible at the point of use.
In PE-backed companies, numbers get defended in boardrooms. Teams need to know how a figure was produced. AI can process inputs and update outputs at speed, but it cannot replace the context that makes a recommendation actionable. The standard is whether someone can stand behind the result without rebuilding it.
The Model That Makes It Stick
The engagements where we see real traction share a common thread. Technical people embedded inside business units, sitting in finance functions, supply chains and operating teams, building against live workflows and iterating with the people who actually do the work. That’s how AI-enabled operating systems get built in a way that sticks, because the people designing them never left the room where the work actually happens.
The results from our own engagements bear that out. In one PE-backed technology company, automating 80% of the close reporting workflow through an agentic system, with finance leaders reviewing and approving outputs through a human-in-the-loop interface, shifted the team from assembling reports to interpreting results. McKinsey’s research finds that finance teams embedding AI robustly spend 20% to 30% less time processing data, redirecting that capacity toward strategy—what happens when AI is built into the workflow rather than on top of it.
The Bottom Line
PE has always won by moving deliberately and fixing what is broken. The firms that apply that same discipline to AI will build a capability that has the potential to compound across every deal. The ones still validating AI outputs in a spreadsheet at month end and calling it transformation will find out what the gap actually costs.
The information provided here is not investment, tax or financial advice. You should consult with a licensed professional for advice concerning your specific situation.
Forbes Finance Council is an invitation-only organization for executives in successful accounting, financial planning and wealth management firms. Do I qualify?
