The AI Jevons Effect: How Automation Could Multiply Lawyers and Accountants
In 1865, an English economist noticed something counterintuitive about efficiency: when machines used coal more efficiently, Britain burned more coal, not less. That observation—now known as Jevons Paradox—reminds us that improvements in productivity can expand consumption. Fast forward 160 years, and the same logic is being applied to artificial intelligence. Rather than erasing demand for professional services, AI may be the engine that enlarges them, producing more lawyers and accountants than we might expect.
From coal to code: the rebound in a digital age
At first glance, AI looks like a classic displacer of routine tasks. Drafting contracts, routine tax filings, due-diligence sifts through documents, and standard audit procedures are all eminently automatable. Give a firm a model that reads documents faster and classifies risks more reliably, and the immediate expectation is fewer billable hours and a leaner workforce.
But efficiency is not destiny. Lower costs per unit of legal or accounting activity change incentives across markets. When the cost of drafting a contract, filing a report, or running a compliance check falls sharply, more organizations and individuals will choose to do those things. Small startups will incorporate more quickly and pursue more sophisticated financing structures. Gig-economy platforms will standardize and scale cross-border engagements. Consumers and businesses will perform a greater number of contractual arrangements, each generating downstream needs for bespoke review, negotiation, dispute resolution, and tax planning. In short, the cheaper the service becomes, the larger the volume of activity—and the more total professional work that follows.
Mechanisms that turn automation into demand
1) Lowering the price barrier multiplies users
When AI reduces the marginal cost of routine legal and accounting tasks, it broadens the market. Tasks that were once prohibitively expensive for small businesses or individuals become affordable. More transactions, incorporations, loans, and contracts happen. Each new transaction carries the potential for exceptions, bespoke clauses, or disputes that require human judgment. The cumulative effect is more baseline demand for professionals to handle the tail of complexity.
2) Complexity breeds specialization
AI handles patterns well, but novel situations push work back to humans. As automation handles the repetitive bulk, human practitioners concentrate on complicated, high-stakes, and ambiguous problems: cross-border tax arbitrage, novel liability claims related to algorithmic decisions, bespoke IP arrangements around AI-generated content, or intricate regulatory interpretations. That concentration increases the value of specialized lawyers and accountants, who in turn expand their services and markets.
3) Regulatory churn and compliance arms races
AI adoption triggers regulatory responses. Governments write new rules for data privacy, algorithmic fairness, model transparency, and liability for automated decisions. Each new regulation increases compliance work: policy interpretation, implementation, audit, and defense. As firms and regulators iterate, the compliance landscape becomes a moving target, sustaining demand for advisers to keep businesses aligned with evolving requirements.
4) New products and markets
AI also creates novel commercial products—automated marketplaces, decentralized finance systems, algorithmic employment platforms—that require legal frameworks and accounting rules. Developers will want standardized terms, tax treatment, and risk-sharing mechanisms. Professionals will design these frameworks, certify their soundness, and litigate or arbitrate disagreements when structures fail or incentives misalign.
5) Liability, trust, and explanation
As AI systems make more consequential decisions, the need to assign responsibility and explain outcomes grows. Who is liable when an algorithmic loan decision causes harm? How does a company prove it exercised due diligence when relying on a third-party model? Answering those questions requires legal reasoning, documentation practices, and forensic accounting. Those are not easily automated away; they are work streams that multiply with more AI deployment.
Historical echoes and modern parallels
Jevons saw that higher furnace efficiency led to cheaper coal service, which increased its use in factories and locomotives, raising overall consumption. Similar patterns have appeared repeatedly with technology: cheaper printing expanded publications and the need for editors and critics; cheaper travel increased demand for hotels, guides, and infrastructure. Today, software reduced the cost of information processing and simultaneously multiplied new products, services, and regulatory obligations. AI promises to accelerate that dynamic.
Consider the rise of web platforms: easier publishing did not diminish the need for journalists, lawyers, or brand managers. Instead, it spawned new roles—content moderating, platform compliance, digital copyright litigation—that scarcely existed before. AI can play the same role for professions: it automates the many, but it also creates the conditions for new, often more complex, many-to-many interactions that require human mediation.
What the professional landscape could look like
Rather than a simple shrinkage, the future is likely to be a qualitative reshaping. Expect three broad trends.
1) Scale and granularity
Automation makes it feasible to provide legal and accounting services at scale and at granular levels. Think micro-contracts, automated audits for small transactions, and instant compliance checks. Firms that can combine AI with human oversight will process vastly greater volumes, creating roles for oversight specialists, exception managers, and customer-oriented legal/accounting technicians.
2) Boutique specialization and high-stakes advisory
Large percentages of routine work will be commoditized, but high-stakes advisory—M&A, antitrust, international tax planning, major litigation—will become even more valuable. Specialists who can translate between deep domain knowledge and AI-driven evidence will be in greater demand.
3) New professions at the intersection
The meeting point of law, accounting, data science, and ethics will spawn hybrid roles: AI auditors, algorithmic compliance officers, and model-liability lawyers. These professionals will craft documentation, design monitoring regimes, testify in disputes, and coordinate cross-disciplinary teams.
Pricing, billing, and the economics of attention
The economic imperative will reshape how services are priced. If routine tasks become nearly frictionless, firms will move from hourly billing toward outcome-based fees, subscriptions, or transaction fees. That pricing shift may increase overall revenue because providers capture value from larger volumes and recurring engagements. Additionally, attention and reputation become scarce resources: who can certify an AI process as trustworthy? Who can offer a defensible, legally robust explanation? These are monetizable services.
Policy and education: preparing for expansion, not elimination
For regulators and educators, the implication is clear: plan for more work, not less. Training programs must emphasize interdisciplinary skills that blend legal reasoning with data literacy. Regulatory design should anticipate that lowering compliance costs may increase the number of regulated activities, requiring scalability in enforcement and dispute resolution mechanisms. Public policy can also shape whether the rebound concentrates benefits or distributes them more broadly—through licensing, certification, and support for small practitioners to adopt AI safely.
Two caveats
First, the rebound is not automatic or uniform. The magnitude depends on elasticities—how responsive demand is to price changes—and on network effects and complementarities that vary across sectors. Some narrow practices may shrink significantly, while entire new subfields expand.
Second, social and ethical choices matter. If AI lowers costs but also erodes quality or fairness, trust will falter and demand may not grow as expected. The path toward expansion requires frameworks that preserve accountability, transparency, and equitable access.
Conclusion: AI as amplifier, not eraser
Jevons’ insight was a caution against assuming that efficiency alone reduces demand. Applied to AI, it offers a different forecast: automation is an amplifier. It reduces friction, multiplies transactions, inflates complexity, and creates new forms of risk and responsibility. Those dynamics will sustain and likely increase demand for lawyers and accountants—though the work they do will change in character.
The provocative takeaway is not that humans will be sidelined, but that their roles will be magnified and reframed. Rather than counting the jobs that vanish, we should be designing systems, training pathways, and regulatory regimes that harness the rebound for public good—so that the expanded market becomes an engine for greater access to justice, better corporate accountability, and more resilient economic infrastructure. In that future, AI does not make professionals obsolete; it makes them indispensable in new and broader ways.
