Ed. note: This article, originally published on April 28, utilizes this LinkedIn post by Gleb Alikhver as a source and originally contained similar language because of a technology error. Portions of this article have been rewritten to remove this similarity. We regret the error.
For decades, the legal industry has operated on a pricing model protected by a comfortable buffer: the gap between what legal work actually costs to produce and what the market has been willing to pay for it. That gap has been sustained by information asymmetry, process opacity, and institutional inertia. It is the foundation on which law firm economics have been built.
AI is collapsing that foundation, and it is doing so faster than most firms or legal departments fully appreciate.
Every industry has structural inefficiencies that sustain its economics. In legal services, the billable hour is not merely a pricing mechanism. It is the operating system of the entire business model. It determines how firms staff matters, how they evaluate associates, how they compensate partners, and how they grow revenue. It is also fundamentally misaligned with the value clients actually receive.
Consider a straightforward example. Two attorneys handle the same type of employment matter. One resolves it in 40 hours. The other takes 120 hours. Under hourly billing, the client pays three times more for the slower attorney, despite receiving the same outcome. The system does not reward efficiency. As most general counsel would acknowledge, it rewards the opposite.
This misalignment has persisted for so long that many in the industry treat it as a law of nature rather than what it actually is: a market distortion that has been too expensive, too invisible, and too entrenched to close. Until now.
Across the broader economy, AI is systematically eliminating gaps between what work costs to produce and what the market charges for it. Recent research from McKinsey & Company and Thomson Reuters has projected substantial productivity compression across knowledge work and professional services as generative AI adoption accelerates.
In financial markets, automated systems have already dismantled inefficiencies that once sustained entire trading desks. The same dynamic is now accelerating across professional services, and the legal industry is squarely in the crosshairs.
AI attacks the economics of legal services on multiple fronts. The most obvious is the production layer. Legal research that consumed hours of associate time can now be completed in minutes. Contract review, document drafting, deposition summaries, and regulatory analysis are all experiencing dramatic compression in production time. When an AI tool can generate a competent first draft of a research memo in minutes, the ten hours historically billed for that task no longer reflect an economic reality.
But the compression goes deeper than speed. AI also eliminates variance in execution quality. A brief produced by AI at 2:00 AM is no different from one produced at 10:00 AM. There is no fatigue, no distraction, no inconsistency. In an industry where variation in human performance has long been absorbed into billable hours without consequence to the provider, this is a direct challenge to the economic model.
AI further commoditizes the information synthesis layer. Law firms have historically charged a premium for the ability to aggregate information from multiple sources and apply judgment across complex fact patterns. When a corporate legal department can run a comprehensive research query across regulatory filings, case law, and internal documents in minutes, the intermediary whose value rests on assembling information loses its pricing power.
Recent updates from corporate legal departments make clear that clients are prepared to act on this trajectory. Meta has rewritten its outside counsel billing rules to treat AI-replaceable work as non-chargeable, and the company has reserved discretion to disallow any invoice entry it concludes could have been produced by a machine. Examples that fall into that category include digests of testimony, form correspondence, and case law surveys on questions that are no longer in dispute.
The pattern extends beyond Meta. Zscaler’s outside counsel guidelines instruct firms not to pass through the time or cost of work product produced by generative AI.
Sophisticated buyers are converging on the same conclusion: when a task is within reach of off-the-shelf automation, an attorney’s hourly rate is no longer the right unit of payment.
Organizations, including the Association of Corporate Counsel (ACC), have also begun publishing model AI-related guidance for outside counsel to legal departments evaluating disclosure, billing, and governance expectations regarding generative AI.
The structural consequences for firms are meaningful. Associate hours have been concentrated for years in exactly the work that commercial AI tools now perform competently: pulling provisions out of contracts, building chronologies, generating first-pass research, and producing record digests of various kinds. As clients begin removing those entries from invoices on principle, the firms that endure will be the ones whose workflows already separate routine production work from the judgment-driven layer where attorney time still belongs. Firms that have not made that separation will see their bills returned with line items struck through.
Hourly billing puts firms in a corner once AI is introduced into the workflow. Saying so transparently invites the client to deduct the time. Staying silent while continuing to bill at full rates produces a growing gap between effort and invoice, which becomes hard to justify under any meaningful scrutiny. There is no good answer available within an hourly framework because the framework itself, not the disclosure question, is what has broken.
The hourly billing model has survived previous waves of technology because those waves were incremental. E-discovery tools made document review faster, but firms adjusted staffing and rates to preserve revenue. Legal research databases reduced time in law libraries, but the billing conversation did not change.
AI is not just another incremental cycle. Capability is expanding across several dimensions of legal production simultaneously, and the cadence of model releases continues to accelerate. The more telling difference, however, is the client-side response. Previous technology cycles arrived without changing the underlying billing conventions. This one is different because corporate clients such as Meta and Zscaler are writing efficiency expectations directly into the agreements governing their relationships with firms. Private contracts between sophisticated buyers and their firms are moving the market in months in ways no regulator could plausibly achieve.
There is a useful parallel from the design and publishing industry. In the mid-1980s, the arrival of desktop publishing software fundamentally disrupted the commercial typesetting business. For decades, producing professional-quality printed materials required specialized typesetting equipment, trained operators, and a production workflow that could take days or weeks. Clients paid for access to that infrastructure because there was no alternative.
When PageMaker and then QuarkXPress arrived, desktop publishing dramatically reduced the cost and complexity of professional print production workflows. A single designer with a Macintosh could produce camera-ready output in hours. The early adopters charged traditional typesetting rates for work done at a fraction of the old cost. For a while, the margins were extraordinary. But within a few years, every design firm had the same tools. Clients realized the output was no longer scarce. Typesetting as a standalone billable service collapsed entirely. The value migrated upstream to design strategy, brand thinking, and creative direction. The production layer became table stakes.
The legal industry is in the early stage of this same arc. Firms using AI to produce deliverables at a fraction of the old cost while still billing at historical hourly rates are enjoying a temporary margin advantage. But that window is closing. As AI tools become universally available and clients develop their own capabilities and OCG enforcement mechanisms, such as Meta’s, the information asymmetry that protects hourly billing will evaporate. The firms and legal departments that recognize this trajectory and act now will be positioned for what comes next. Those that continue to operate as if hourly billing is permanent will find themselves on the wrong side of a rapid repricing.
The good news for Legal Operations professionals is that the alternative to hourly billing is not hypothetical. Value-based pricing (“VBP”) has been the standard in virtually every other major professional services industry for decades. Management consulting firms, accounting firms, and investment banks all moved away from hourly billing long ago. They price on deliverables, outcomes, and defined scopes of work. The legal industry has been the last holdout.
Under a properly structured value-based pricing model, clients pay fixed fees tied to specific tasks, phases, and deliverables. The conversation shifts from effort to outcomes. Budget predictability improves dramatically. Invoice review, which in some legal departments consumes 10 to 20 percent of in-house attorney time, is eliminated entirely. And total legal spend typically drops by 20 to 50 percent.
VBP also resolves the AI disclosure paradox that hourly billing creates. When a firm is paid a fixed fee for a defined phase of work, it does not matter whether the firm used AI, associates, or a combination of both to produce the deliverable. The client is paying for the outcome, not the input. The firm is incentivized to be efficient, to deploy AI where it adds value, and to apply attorney judgment where it matters. There is no conflict between disclosure and compensation.
The transition to VBP does not require firms to take on unlimited risk. Properly structured fixed-fee arrangements use per-occurrence pricing for unpredictable activities like depositions or motions, phased pricing that reflects the natural progression of a matter, and defined scopes that make the economics clear to both sides. This is not a capped-fee arrangement, which still requires hourly billing and invoice review. It is a fundamentally different approach to pricing legal services, based on the value delivered rather than time spent.
AI does not eliminate the need for lawyers. It eliminates the need for a particular type of legal work to be performed by lawyers as it has always been done. The value does not disappear. It migrates upstream.
When AI collapses the cost of legal research, the value shifts to judgment, strategy, and client counseling. When AI automates contract drafting, the value shifts to deal structuring, negotiation, and risk assessment. When AI handles the production layer of litigation, the value shifts to case strategy, courtroom advocacy, and settlement judgment. This pattern is predictable and consistent: the new value is always closer to judgment, taste, and relationships, and further from production, execution, and information retrieval.
The economics of that migration only work if the pricing model changes along with the work. You cannot price upstream judgment on an hourly basis and expect the market to function rationally. The attorney who resolves a matter with a single well-placed phone call delivers enormous value. Under hourly billing, that value generates a fraction of the revenue that a drawn-out process would. VBP corrects this by paying for the outcome, not the clock.
The growing tension between generative AI and the traditional leverage-based law firm model has also been increasingly discussed across the legal industry and academic literature.
The pace of AI development is accelerating. Major model releases are now quarterly, with each release expanding the frontier of what can be automated. The gap between firms that have adopted AI and those that have not is growing. Meanwhile, the gap between legal departments that have moved to VBP and those still mired in hourly billing is growing even faster.
More importantly, corporate clients are not waiting for firms to adapt. Meta’s OCG update is not an isolated event. It is the leading edge of a wave. As more legal departments adopt their own AI-specific billing provisions, firms that have not restructured their economics will face a choice between disclosing AI use and accepting reduced revenue or remaining silent and hoping clients do not notice. Neither option is sustainable under the hourly model.
For Legal Operations professionals, this is not a future problem. It is a present-tense strategic decision. Every month spent reviewing hourly invoices for work that could be priced on a fixed fee basis is a month of wasted in-house attorney productivity. Every engagement structured on an hourly rate is one where the client bears all the risk, absorbs all inefficiency, and has no budget predictability. The firms that will thrive over the next five years are the ones that embrace both AI-driven efficiency and value-based pricing. The firms that cling to the billable hour will find their economics hollowed out as clients like Meta simply stop paying for the work that AI can do.
Ken Callander is Managing Principal of Value Strategies LLC, a consulting practice that advises corporate legal departments on outside counsel pricing strategy. He previously served as Head of Legal Operations at Uber Technologies. He is a Certified Pricing Professional and holds a degree in Physics from Stanford University.