Written By: Don Jaycox and Dan Safran of Unbiased Consulting, LLC
Featured in Law.com
Generative AI is no longer a futuristic concept for law firms. It is here, it is powerful, and it is reshaping the business of law. Yet, despite the urgency, many firms remain caught in a cycle of experimentation—running pilot after pilot with different tools but failing to move beyond testing. The result: wasted resources, skeptical attorneys, and little measurable return on investment.
The Pitfalls of “Pilot Purgatory”
The most common mistake we see is a tool-first approach. Firms acquire licenses, set up pilots, and hope to discover use cases along the way. Without clarity about outcomes or integration into actual workflows, these efforts inevitably stall. Lawyers tire of testing tools that don’t improve their day-to-day work, and leadership loses patience with investments that don’t pay off.
This “pilot purgatory” erodes credibility. Teams risk being viewed as chasing trends rather than delivering value, while attorneys become more resistant to future change initiatives.
The smarter path is to begin with ways of working, not tools. That means asking:
- What are the protocols, workflows, and processes that drive the practice?
- Where are the pain points for lawyers and staff?
- How does work move from associate to partner to client?
- What outcomes do we want—efficiency, risk mitigation, new service models?
Only once those answers are known and the future way of working is clear should firms choose tools. Technology should support the workflow, not dictate it. Without this foundation, AI risks amplifying bad processes instead of improving them as technology acts as a magnifier—it makes a good process better, and a bad process worse.
The Risk Is Real
Generative AI carries unique risks if implemented haphazardly. The most visible is the threat of hallucinated citations and inaccurate outputs. Courts in the U.S. and abroad have already sanctioned attorneys for improper use of AI, with penalties reaching as high as six figures.
But reputational damage can be worse than fines. Clients expect rigor, accuracy, and ethics. A single AI misstep in a filing or transaction can jeopardize trust built over years. Structured guardrails—review protocols, clear rules of use, and defined oversight—are essential for protecting both the firm and its clients.
Beyond risk, AI exposes a fundamental economic trap for law firms. In most industries, efficiency is a direct path to higher profit: costs go down, margins go up. In law, the opposite often holds true. When tasks that once filled an hour are finished in minutes, clients expect lower bills, and ABA rules prevent firms from charging for time not spent. The paradox is stark: the more efficient a firm becomes, the less revenue it generates. Instead of expanding margins, efficiency erodes them.
The way out lies in rethinking the model, shifting from time-based billing toward value-based pricing, outcome-driven fees, and new service models that reward results, not hours.
This disincentive to innovate is one reason adoption lags. But it is also why firms must rethink the business model. Value pricing and alternative fee arrangements (AFAs) offer a solution. Instead of charging by the hour, firms can price based on outcomes, expertise, or packaged services.
Predictions that AFAs would make up half of firm revenue never materialized; most firms hover around 15%. But generative AI may finally accelerate the shift, as clients increasingly demand predictable costs and balk at paying for inefficiency.
Unlocking Opportunity
Too often, firms view AI only through the lens of efficiency. The bigger opportunity is new models of service delivery. For example:
- Reducing write-offs in due diligence or research by applying AI to areas clients resist paying for.
- Streamlining nonbillable tasks like time entry, email synthesis, and pro forma reviews, freeing lawyers for higher-value work.
- Productizing services, turning repeatable work into subscription-based or fixed-price offerings.
- Expanding capacity by allowing associates to handle more matters without compromising quality.
These models shift the conversation from “how do we cut time?” to “how do we increase value?”
In many firms, resistance to change is as much a barrier as technical complexity. Lawyers accustomed to traditional practices may distrust new tools, while leadership may hesitate to make investments without guaranteed ROI.
This is why leadership engagement is non-negotiable. Successful firms treat AI as a business initiative, not an innovation experiment or an IT project. They align practice leaders, executives, and professional staff around clear goals. They invest in training, change management, and governance. And they frame AI adoption in terms lawyers understand: client service, risk reduction, and profitability.
While there is no one-size-fits-all roadmap, but we see two effective approaches:
- Strategy-Led Discovery: A top-down, holistic effort to assess opportunities, align leadership, and build a firmwide roadmap tied directly to business outcomes.
- Design Thinking-Led Ideation: A bottom-up process that engages lawyers and staff in identifying problems, generating solutions, and building grassroots buy-in.
Both approaches share a common thread: they begin with business goals and workflows, not with technology pilots.
One Thing is Clear
One thing is clear: generative AI is here to stay. The firms that succeed will not be those running the most pilots, but those that build deliberate strategies, align leadership, and embed AI into the way work gets done.
Stop piloting for the sake of it. Start with business outcomes. Redesign processes and guardrails. Rethink pricing models. And then, with clarity of purpose, choose the tools that enable the future of legal work.
About the Authors
- Dan Safran is President of Unbiased Consulting.
- Don Jaycox is the former CIO of DLA Piper and a Senior Consultant at Unbiased Consulting.