Welcome to Attorney Intelligence, where we break down the biggest advancements in AI for legal professionals.

This week, we consider the areas in which law firms are consistently making use of AI, and others where high expectations have been challenged by reality.

Expectations in 2024 don’t match the realities of 2025

When the legal industry was first introduced to tools that promised to harness the power of generative AI to attorneys’ benefit, many predicted a revolution in the nature of legal work. Last year, well into this wave, Bloomberg Law conducted a survey to gauge expectations.

The Bloomberg survey set out to map attorneys’ expectations in 2024 of the utility of AI tools against their perceptions of those same tools in 2025. Expectations fell short of reality for each of the surveyed work functions, although the divergence between the two factors varied significantly. Workflow automation was the greatest disappointment, with nearly 80% of surveyed firms expecting AI to increase automation in 2024, and less than 40% finding that to be the case the following year. Relatedly, over 60% of surveyed firms expected that adoption of AI tools would lead to increased time being spent on “higher level” tasks—but this year, only 30% of firms found that to be the case in retrospect.

The Bloomberg survey suggests that—as of 2025—the adoption of AI legal tech products has in many ways underwhelmed, and that the industry as a whole is resetting its expectations. The difficulty of accurately gauging expectations aside, however, a more nuanced reading paints a different picture: AI is transforming different areas of the legal industry at different rates.

AI is penetrating different areas of legal work at different rates

Anticipated erosion of the billable hour model has not materialized

The billable hour model forms the backbone of law firm revenues. When AI enhanced-productivity tools first came on the scene, some lawyers feared that significant productivity gains stemming from AI adoption would quickly eat into their firm’s bottom line. To compensate, Alternative Fee Arrangements (AFAs) would proliferate as law firms attempted to safeguard revenue and recoup investment costs associated with AI.

In 2024, 40% of firms of law firm lawyers surveyed by Bloomberg expected that AI would lead to an increase in AFAs over the following year. What did the survey find a year later? Only about 10% witnessed an increase in AFAs—one of the greatest relative and absolute differences between expectations and reality across all work functions. A Harvard Law School analysis of the impacts of AI adoption on law firms’ business models found that the dominant expectation among lawyers today is that the total number of hours worked would be similar to—or even greater than—the hours currently worked by lawyers without AI. Rather than fundamentally changing the business model of firms, 90% of the firms interviewed agreed that additional time would likely be invested in improving the “quality of service” from outside counsel—and their clients, agreeing with this expectation, are “comfortable” with the current fee arrangements.

In the long term, firms are continuing to develop their capabilities around alternative pricing agreements. But none of the surveyed firms believed that their capabilities were sufficiently advanced to accurately scope and price work—meaning that for the foreseeable future, the billable hour model remains dominant.

AI for client work is under-penetrated while AI for internal work and firm analytics is relatively strong

While the scope of AI adoption broadly is expanding, law firms that are actively leveraging legal tech for client work remain in the minority. In another recent Bloomberg Law survey, the share of law firms that reported using generative AI for client work was just north of 40%, far less than the reported share for internal firm use. In contrast, over 60% of surveyed firms were using data analytics to understand their own business, and roughly 60% were using generative AI for internal use only. A smaller subset of those firms is likely leveraging AI to inform resource management decisions at the case matter level; case or matter strategy development was the second most common application of AI in the American Bar Association’s 2024 Legal Technology Survey Report, after legal research.

The relatively fast pace of legal tech adoption for internal firm analytics can be partly explained by AI’s facility in organizing and parsing vast stores of data, including data from native case management systems. It also points to an emerging priority for tech-focused law firms: Generating better intelligence to inform firm strategy. The prospect of greater efficiency and flexibility in competing for business will likely continue to drive AI adoption in this space.

Legal Bytes

Anthropic pays out in largest ever copyright settlement - Four sitting US Supreme Court justices and certain presidents are eligible to collect settlement payments after a court ruled that the AI giant had violated copyright laws (Bloomberg Law).

Law school experiments with AI jury - The University of North Carolina law school has conducted simulated trials featuring AI jurors (Reuters).

Aderant to acquire Virtual Pricing Director (VPD) - Aderant announced that it intends to acquire the London-based pricing platform to marry firm data with pricing strategies developed by VPD (Stock Titan).

Events

I’ll be speaking on the panel “From Metrics to Mastery - The Rise of Legal Performance Dashboards” at Future Lawyer USA tomorrow (29 October). If you are attending and interested in learning more about what performance indicators firms are tracking and why, stop by!

Looking to improve your firm’s efficiency?

Time entry is the foundation of law firm data strategy, but currently, few tools exist to allow for meaningful analysis of time entries. Fortunately, PointOne is designed to turn time sheet data into actionable insights for leading law firms.

Book a demo to see how PointOne can stop revenue leakage and drive insights across your practice.

Until next week,

Katon