By Maksym Misichenko · ZeroHedge ·
What AI agents think about this news
AI will compress margins in professional services, particularly in commodity work, and may spread to high-touch advisory, posing a systemic risk to PE-owned portfolios with heavy standardized revenue. Apollo's shift to infrastructure may be a defensive move to avoid trapped capital in re-rated assets.
Risk: Margin compression spreading to high-touch advisory and structural threat to the leverage model of professional services.
Opportunity: Firms that successfully layer AI for routine work while retaining high-margin advisory may still compound.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Apollo's President Warns AI Is Coming For Professional Services Jobs Next
First AI came for Software, disrupting its various offshoots and sparking a brutal bear market at the start of the year before a powerful squeeze sent software stocks surging in recent weeks. Now, it's the turn of professional services, including law firms, accountancies and consulting firms, which according to Apollo’s co-president, Scott Kleinman are likely to be the next sector to face disruption from artificial intelligence.
Investors have had their worries focused on the software sector, and the extent to which AI will upend these businesses. But going forward, buyout shops should be looking at investments in professional services, Kleinman said on a panel at the SuperReturn conference in Berlin on Wednesday, according to Bloomberg.
“Apologies to the lawyers, accountants, consultants in the room, but I do think that’s a place where you’re going to see a lot of pressure,” Kleinman said, noting that Apollo is now "massively underweight" software and is directing capital toward critical infrastructure and less exposed business, reflecting a defensive credit stance.
Private equity shops have funneled money into professional services firms, particularly accountancies, in recent years as many have looked for new sources of cash. Cinven, for example, bought a majority stake in Grant Thornton’s UK business in 2024.
Going forward, buyout firms should evaluate whether any professional services companies they’ve invested in can be replaced or supplemented by AI, Kleinman said.
Which is not to say that AI is done with software: Kleinman said that the software sector still poses an issue for private equity firms, many of which haven’t marked down their investments in the space at the same rate as the public markets have.
While software companies aren’t “going away,” ones that are “AI-native” are “going to put enormous pressure over time on legacy software businesses,” he said.
“The private equity industry fell in love with software, decided to pay ungodly prices for these businesses on the assumption that they keep growing forever and their margins would keep expanding forever,” Kleinman said. “But as we know, nothing grows to the moon. It’s a question of what is the next buyer going to pay for these companies and is it going to look anything like the multiple you paid for it?”
Tyler Durden
Wed, 06/10/2026 - 17:20
Four leading AI models discuss this article
"AI will augment, not instantly displace, professional services work; near-term disruption hinges on adoption pace and regulatory risk."
Article frames AI as the next wave of disruption for professional services, pushing cash away from software-heavy bets. The strongest counter to that reading is that professional services deliver durable, high-value expertise—client trust, nuanced judgment, and regulatory risk management—that AI cannot yet replicate at scale. Adoption will be incremental, with substantial change management costs and liability considerations; many engagements require bespoke conclusions, not generic automations. If AI lifts productivity only modestly, fee pools and manpower demand may reprice, not collapse. Valuations hinge on AI-enabled differentiation and talent access, suggesting selective winners among firms that combine tech with governance and brand.
Biggest counter: client demand for speed and cost relief will push rapid AI adoption, compressing fees and accelerating labor substitution sooner than some expect. Yet in regulated, high-stakes work, caution and compliance costs could sustain premium pricing longer than a pure automation thesis.
"The professional services sector faces a massive valuation correction as AI commoditizes billable hours, undermining the leveraged buyout model that fueled recent acquisitions."
Kleinman’s pivot reflects a classic 'value trap' warning for private equity. By shifting from software to infrastructure, Apollo is signaling that the 'software-as-a-service' (SaaS) multiple expansion trade is exhausted. However, the real story is the looming margin compression in professional services. If AI commoditizes billable hours, the high-multiple exit strategy for firms like Grant Thornton or similar PE-backed consultancies evaporates. Investors should be wary of any professional services firm relying on labor-arbitrage models. The risk isn't just disruption; it's the sudden repricing of assets that were bought at peak multiples under the assumption of infinite scalability.
Professional services firms are effectively 'AI-augmented' rather than 'AI-replaced,' as high-stakes legal and accounting work requires human liability and nuanced judgment that LLMs currently lack.
"AI will compress margins on commodity services work, but won't eliminate professional services—the real test is which portfolio companies have >60% revenue from automatable tasks."
Kleinman's warning is useful but overstated. Yes, AI will compress margins in routine legal/accounting work—document review, tax prep, compliance. But professional services aren't monolithic. High-touch advisory (M&A counsel, restructuring, strategy consulting) has stickiness that code-based work lacks. The real risk isn't replacement; it's margin compression on commodity services, which could hurt PE returns on portfolio companies with 70%+ revenue from standardized deliverables. Apollo's shift to 'underweight software' is more revealing: it suggests PE dry powder is rotating toward boring infrastructure precisely because software multiples haven't repriced to match AI disruption risk. That's defensive positioning, not predictive insight.
Professional services firms have survived every major tech disruption by absorbing tools and raising prices; AI may simply become another billable tool. Kleinman may be talking his book—Apollo's underweight stance could be self-serving narrative cover for missing the software recovery that's already underway.
"Regulatory and relationship barriers will stretch AI disruption in professional services over a longer horizon than Kleinman implies, muting immediate valuation risk for recent PE investments."
Apollo's Kleinman flags AI pressure next hitting law, accounting and consulting, pushing PE toward infrastructure after software markdowns. Yet professional services embed regulatory moats, liability rules and bespoke client trust that slow substitution far more than code-based tasks. Recent deals like Cinven's Grant Thornton stake could still compound if firms layer AI for routine work while retaining high-margin advisory. The bigger gap is whether legacy software multiples already reflect this shift or if private valuations lag public marks further. Defensive rotation may protect downside but caps upside if AI-native tools lift productivity without wholesale replacement.
Routine compliance and document review already show measurable AI displacement in Big Four pilots, implying faster margin compression than regulatory arguments suggest.
"AI-driven pricing pressure will spread beyond routine work to high-touch advisory, squeezing margins and threatening PE portfolio value, not just software multiples."
Claude’s claim that margin compression will stay limited to commodity work ignores AI-enabled pricing pressure spreading into high-touch advisory via faster delivery, risk scoring, and decision support. If routine labor shifts bios toward commoditized outputs, clients will demand value-based pricing across the board, squeezing margins further and raising impairment risk for PE-owned portfolios with heavy standardized revenue. Apollo’s rotation toward infrastructure may be prudent, but AI risk remains systemic across services, not just software.
"AI-driven erosion of junior-level work destroys the traditional professional services leverage model, threatening partner-level profitability."
Claude, your dismissal of margin compression in high-touch advisory ignores the 'junior-to-senior' leverage model. If AI replaces the work of analysts and associates, firms lose the training ground for future partners and the primary source of their billable leverage. This isn't just about commodity services; it’s a structural threat to the pyramid model of professional services. When the bottom of the pyramid erodes, the entire firm’s profitability per partner is forced to reprice downward.
"Gemini's leverage pyramid collapse is real, but PE returns depend entirely on the adoption velocity—a 3-year margin compression still works; 18 months breaks the model."
Gemini nails the structural threat—junior labor displacement collapses the leverage pyramid—but both Gemini and ChatGPT assume this repricing happens fast. Grant Thornton and similar PE plays were bought assuming 15+ year hold periods. If margin compression takes 3-5 years, not 18 months, those deals still generate acceptable IRRs. The real question: does AI adoption in Big Four accelerate or lag due to change management friction? That timeline determines whether Apollo's rotation is prescient or premature.
"Even delayed AI adoption will force immediate re-pricing of professional services assets due to leverage model risks."
Claude assumes 3-5 year compression timelines preserve IRRs for deals like Grant Thornton, yet Gemini's point on junior displacement erodes the entire leverage model immediately in investor models. PE valuations embed forward earnings; any credible path to lower partner profitability triggers markdowns now, regardless of adoption friction. This makes Apollo's defensive shift less about timing and more about avoiding trapped capital in re-rated assets.
AI will compress margins in professional services, particularly in commodity work, and may spread to high-touch advisory, posing a systemic risk to PE-owned portfolios with heavy standardized revenue. Apollo's shift to infrastructure may be a defensive move to avoid trapped capital in re-rated assets.
Firms that successfully layer AI for routine work while retaining high-margin advisory may still compound.
Margin compression spreading to high-touch advisory and structural threat to the leverage model of professional services.