ANNOTATION: This question focuses on friction. After identifying what’s expected, we now ask what’s preventing progress—what most constrains your ability to adopt or scale GenAI in legal service delivery.

You are asked to identify a single “primary obstacle”—the most immediate, structural barrier your organization faces—and then to rate how serious the other listed challenges are. We recognize that every organization will face some combination of these issues, but the exercise of naming one as primary helps clarify what’s truly gating acceleration, not just complicating it.

We also recognize the irony: You’re unlikely to have a fully objective view of your own obstacles, much less their relative weight. Many of these factors overlap: Unclear ROI may be the surface expression of cultural resistance—i.e., the cultural resistance manifests as complaints about the lack of a clear business case. The categories are not meant to be neat or mutually exclusive; they’re designed to capture recurring patterns of friction that shape GenAI progress across the ecosystem.

GenAI Not There Yet. For some, the primary barrier is the technology itself—its current limits in accuracy or reliability within legal contexts. Even promising tools often fail to meet the profession’s threshold for confidence and defensibility. This can breed caution: the sense that GenAI is “almost but not yet ready” for high-stakes use. The resulting skepticism is rational, not regressive; it reflects the uniquely high reliability standards of legal work.

Search-to-Implementation Speed/Costs. Identifying, testing, and integrating tools takes time and money. Pilots are expensive, success rates are uneven, and lessons learned are hard to scale. Many organizations struggle to move from experimentation to operationalization because discovery cycles consume disproportionate bandwidth. For some, the obstacle is not so much resistance as it is simple capacity constraint: too many variables, too few hands.

Unclear ROI/Business Case and Budget Constraints. Even when enthusiasm is high, progress stalls without a credible economic case. Many leaders hesitate to fund initiatives that can’t demonstrate clear returns or client impact. The result: Projects pause not for lack of belief but for lack of a budget-backed rationale.

Change Management and Cultural Resistance. Organizations don’t adopt technology; people do. Time pressure, entrenched habits, and fear of obsolescence can all slow adoption. Lawyers in particular are conditioned to limit downside risks, prizing precedent and precision over experimentation. “Change management” in this context is about alignment, not attitude: creating space, safety, and incentive for behavior change at scale.

Talent and Skills Gap. There’s a shortage of professionals who can bridge the gap between technology and practice. Knowing what GenAI can do is not the same as knowing how to make it work inside a matter, a team, or a workflow. The challenge is not just hiring technical talent—it’s cultivating applied literacy among lawyers, operations, and technologists alike.

Data Readiness and Quality. GenAI depends on what it can access and trust. In most legal organizations, knowledge is trapped across systems, formats, and minds. Even the best models underperform when data is fragmented, inconsistent, or poorly tagged. This obstacle is foundational: Until knowledge is structured and findable, GenAI remains limited to shallow use cases.

Legacy Systems and Integration Difficulties. Most organizations operate within complex, interdependent IT ecosystems. Introducing GenAI often means confronting technical debt. Integration hurdles make innovation feel like surgery: invasive, slow, and risky.

Information Security and Client Confidentiality. Legal work involves sensitive, privileged, and regulated information. Many organizations are constrained not by unwillingness but by duty. Even if tools are secure in principle, risk tolerance varies widely. Until model training, data isolation, and auditability mature, this remains a universal gating concern.

Regulatory Volatility and Professional Responsibility Ambiguity. Rules are evolving faster than clarity. Bar guidance, data residency laws, and AI governance frameworks differ by jurisdiction and change frequently. Providers are caught between undercompliance risk and overcompliance paralysis. The absence of consistent standards breeds confusion about what is “safe enough” and consternation about how quickly that can change.

We recognize that these obstacles overlap. They’re not meant to be silos; they’re interdependent symptoms of a market still forming its norms. You’re not expected to have perfect clarity on what’s blocking progress, but you’re making decisions based on your best interpretation—and that’s what we’re capturing here.

Finally, if your main challenge isn’t reflected here—or if the interplay among them is more complicated—use the optional commentary box to elaborate, clarify, or add missing context.