Zero AI Payoff Outside Tech Spells Trouble for Law Firm Profit Models

By Grace Kim

Artificial intelligence has delivered "zero financial payoff to corporate America outside of tech." That blunt assessment comes from Torsten Slok, chief economist and partner at Apollo Global Management, in a recent analysis that lands at an awkward moment for the legal profession.

His argument arrived just before ClioCon and the release of Clio's Legal Trends Report. That report, together with an earlier billing study from 8am, showed billable hours going up rather than down — despite widespread predictions that AI would shrink them. Joe Patrice of Above the Law examined that paradox in a recent piece titled "AI Not Only Failed to Kill the Billable Hour, It's Looking Like Its Bodyguard."

Who is making the claim?

Slok is no fringe voice. He spent 15 years at Deutsche Bank before joining Apollo. He previously worked at the International Monetary Fund and the OECD. He is known for challenging conventional wisdom, and his data point is stark: revenue per employee and profit margins have remained largely flat among S&P 500 companies outside the Magnificent Seven — Apple, Microsoft, Nvidia and their peers.

"There are still no signs of AI boosting productivity growth," Slok wrote. In his words: "With no impact on the bottom line, mainstream corporate America is pulling back and starting to view token spending not as a source of growth but as a cost center that weighs on margins." Tokens are the meter that runs each time someone uses an AI tool.

The practical consequence: mainstream corporate buyers may decide they do not need the new AI tools. And they may stop paying for them.

What does this mean for law firms?

Stephen Embry, writing at Above the Law, draws out the implications for practitioners. AI can boost law firm profits by automating back-office tasks. But that mostly cuts costs. It affects revenue only indirectly, by giving revenue producers more time to bill hours — which may explain what Clio and 8am observed.

Beyond that, firms hit a structural paradox. Every billable hour AI saves is an hour of lost revenue. There is no getting around it.

As token costs rise alongside the price of sophisticated AI tools, firm management may ask a legitimate question: why buy technology that reduces profits? Unlike corporate America, where AI could conceivably raise revenue per worker, the predominant law firm business model may never see that benefit.

Will clients force the issue?

Clients could still reshape the market. Embry identifies three pressure points:

• Clients may demand to see AI use reflected in their bills
• They may move to firms that demonstrate it
• They may bring more work in-house

But here Slok's deeper hypothesis intervenes. His analysis states: "AI may be a great calculator, but in business, intelligence is mostly judgment. Judgment depends on context, and the answer varies with which human or which model you ask." Crucially, he adds that "AI does not understand accountability, trust, culture, unwritten rules, ethical trade-offs or fairness, all of which shape reputational risk and ultimately drive revenue growth and profit margins."

Why judgment protects legal fees

Law runs on judgment more than almost any other business. It is the ability to predict what a client will want before the client knows it. It is predicting what an adversary will do — telling a client at what point the other side will seek settlement, and being right. It is knowing the difference between a fair, achievable result and one that will never happen.

That has pricing consequences. Judgment becomes more valuable when everyone else thinks with AI. As its price goes up, the savings from AI go down. Since so much of law hinges on judgment that AI cannot provide, even widespread AI adoption would do little to reduce the value and overall cost of legal services. And if AI does not reduce the client's cost, the client has little reason to demand it — or for firms to buy the tools at all.

Even if the profession masters alternative fees, Embry argues, a premium for judgment will remain built into the fee.

The takeaway

The Clio study shows AI has not yet produced substantive-side savings comparable to its back-office gains. If analysts like Slok are right, substantive adoption may be slow, and wholesale fee reductions may never arrive in the form currently predicted.

AI may still transform legal work — just not in the ways the gushing forecasts describe. Not unless it replaces judgment, and one human's ability to understand another. That, Embry concludes, is what practicing law is all about.