But scale may matter more, not less, in the accounting and legal markets.
By CPA Trendlines Research
The same technology that promises higher margins could weaken the billable-hour economics that made professional services so attractive.
MORE Private Equity
Private equity’s rush into law and accounting is running into a new question: What happens to a roll-up strategy built on professional labor when artificial intelligence starts doing more of the work?
At the SuperReturn private capital conference in Berlin, private equity and credit executives warned that generative AI could disrupt law, accountancy and other asset-light advisory businesses that bill by the hour, according to the Financial Times.
Kevin Marchetti, chief investment officer and head of U.S. direct lending at Man Group, told the FT that AI risk extends well beyond software and could affect “claims auditing, billing automation, proxy voting management or legal services.”
The warning lands at an awkward moment for professional services. Buyout firms have spent the past five years moving deeper into accounting, tax, advisory and legal services, drawn by recurring revenue, fragmented markets, aging owners, sticky clients and labor-intensive delivery models.
The accounting deals are no longer small or experimental.
Grant Thornton Advisors and Grant Thornton Ireland closed a New Mountain Capital-backed transaction in January 2025 to create a multinational platform of roughly 12,000 professionals in more than 50 offices, according to Grant Thornton.
Baker Tilly announced in February 2024 that Hellman & Friedman and Valeas Capital Partners were making what the firm called the largest U.S. advisory CPA private equity transaction to date.
Blackstone announced in January 2025 that it would acquire New Mountain’s stake in Citrin Cooperman Advisors, a professional-services firm with more than 450 partners and 2,800 professionals.
The legal-services market is drawing similar attention, although law-firm ownership rules make direct investment harder in most U.S. jurisdictions. Advisory firm Arrowpoint reported 90 European legal-services deals in 2025, up more than 23% from 2024, and says private equity-backed platforms were creating more competition for quality independent firms.
Arrowpoint reports 19 legal-services deals in the first quarter, almost unchanged from 20 in the prior quarter, and PE Hub reported that AI uncertainty was slowing some legal-services deal execution even though private equity interest had not disappeared.
The issue is not whether AI can make law and accounting firms more efficient. Firms and vendors are already saying it can. The issue is whether the efficiency belongs to the firm, the client, the employee, the software provider or the private equity sponsor.
Bloomberg Tax reports that top U.S. accounting firms are developing new pricing models as they invest billions of dollars in AI tools that could slash billable hours. The same story quotes PwC U.S. Tax Leader Krishnan Chandrasekhar saying, “Time’s becoming less and less of relevance.”
And it quotes Fore LLC CEO Michelle River warning that firms that do not adopt “a more worth-based approach” may need to add more clients to replace lost hours.
If a tax advisory report once took 20 hours and generative AI plus professional review cuts the work to 10 hours, the client may still receive the same report, or a better one. But the revenue math changes if the firm is still selling hours. RSM’s Sergio de la Fe put the question directly in Bloomberg Tax: “Is that all margin for me or is that all price reduction for our client?”
Legal buyers are making the same argument in plainer language. Business Insider reports that Jeff Bleich, Anthropic’s general counsel, told an American Bar Association audience, “I don’t think the billable hour is the solution.”
Bleich says AI will eliminate some of the “tedious work” that historically supported large legal teams.
Damon Hart, Liberty Mutual’s top lawyer, told the same panel, “The value is no longer you putting in time. The value is your strategy, your results.”
For private equity, that cuts to the underwriting model.
Professional-services deals traditionally depend on a mix of revenue stability, partner succession, cross-selling, add-on acquisitions, back-office consolidation, offshore delivery, rising realization rates and operating leverage. AI can help with several of those levers, but it can also weaken the relationship between staff hours and revenue.
In accounting, the exposure is broadest in repeatable, document-heavy and research-heavy work.
Tax preparation, K-1 processing, provision support, audit preparation, research memos, client accounting services, due diligence and compliance-heavy advisory work all contain tasks that AI systems can summarize, draft, classify, reconcile or review.
In law, the obvious targets include document review, contract analysis, legal research, first-draft memos, diligence, discovery and routine correspondence.
The private equity counterargument is equally important. PE-backed platforms may be better positioned than traditional partnerships to fund AI at scale, rebuild workflows, centralize data, standardize processes and absorb the cost of enterprise systems.
Grant Thornton Advisors says it will invest $1 billion over three years in AI tools and technology across its multinational professional-services platform. The firm says the platform includes more than 13,500 professionals in 60 multinational offices and that it is rolling out Microsoft 365 Copilot after a 400-person pilot generated productivity gains.
Grant Thornton’s AI announcement explicitly tied the technology push to its New Mountain-backed platform.
Tom Puthiyamadam, Grant Thornton Advisors’ managing partner for advisory services in the U.S., says the firm was “building a full-service platform amplified by AI and advanced technologies.”
New Mountain managing directors Andre Moura and Nikhil Devulapalli says they supported the AI investment as Grant Thornton built “one of the strongest technology-enabled platforms in the accounting profession.”
But scale may matter more, not less, in an AI market.
Large platforms can spread technology costs across thousands of professionals. They can centralize client data. They can negotiate with software vendors. They can impose workflow standards across acquired firms. They can buy specialist technology businesses. They can run pilots, measure savings and redeploy staff. Smaller firms may use the same public AI tools, but they may not have the capital, data discipline or operating structure to turn those tools into margin.
To be sure, many accounting roll-ups are still stitching together acquired firms with different practice-management systems, client records, billing habits, partner expectations and local cultures. AI systems are only as useful as the data and workflows around them.
A roll-up that has not standardized its client data, engagement letters, pricing logic, quality controls and delivery process may discover that AI exposes fragmentation faster than it fixes it.
The regulatory overlay is also getting heavier. The International Ethics Standards Board for Accountants has launched a dedicated workstream to analyze ethical and independence issues related to private equity investment in accounting firms.
IESBA said the workstream would assess whether formal standard setting is needed for alternative practice structures, including structures involving private equity. The same board says it would also develop new non-authoritative guidance on technology and artificial intelligence.
The PE-CPA model depends on separating attest services from nonattest advisory, tax and consulting work.
AI complicates that structure because quality, independence and economics are intertwined. If AI raises margins in tax and advisory work, sponsors may press for faster deployment and broader automation. If AI is used in audit or audit-adjacent workflows, regulators and clients may ask whether quality controls, human review and independence safeguards have kept up. If AI reduces the staff hours needed for compliance work, partners may ask whether to cut prices, protect margins, redeploy people into advisory work or raise output expectations.
The legal market faces its own version of the same problem.
Most U.S. states still restrict nonlawyer ownership of law firms, which limits the direct CPA-style roll-up model. Investors have instead looked at legal-services businesses, alternative business structures where permitted, legal operations, document-management companies, claims services, e-discovery, litigation support and law-firm-adjacent platforms. But AI’s pressure on document-heavy legal work may change which assets look safe and which look exposed.
AI is changing the unit of value. The old unit was time multiplied by rate, supported by partner leverage and staff leverage. The new unit is more likely to be workflow, data, client outcome, risk reduction, speed, specialization and pricing power.
That is why the same technology can be both a threat and an opportunity. A firm that uses AI to complete a fixed-fee tax project faster may expand margin. A firm that uses AI to complete hourly work faster may reduce its own revenue. A firm that trains staff, cleans data and reprices engagements may become more valuable. A firm that simply buys software licenses may create savings that clients quickly demand back.
For CPA firms, the hardest question may be cultural rather than technical.
The profession has spent decades teaching owners, managers and staff to track time, manage realization and measure productivity by hours. AI pushes firms toward a different system, where the client pays for certainty, speed, insight and outcome rather than recorded effort.
For private equity owners, the hard question is financial. If AI raises margins, PE-backed professional-services platforms could become stronger. If AI turns billable work into price pressure, some assets may be worth less than their acquisition models assumed. If AI requires a large upfront technology investment before revenue models change, sponsors may have to fund a transition that depresses near-term cash flow. If AI favors the biggest platforms, the consolidation wave could accelerate.
The first phase of private equity’s accounting bet was about succession, scale and capital. The next phase will be about pricing, data and workflow control.
Private equity did not necessarily buy the wrong industry. It may have bought the right industry just as the economics of professional labor started to change.