Who Owns the Work When AI Crosses Every Department?

A preview of The Convergence Era panel at the Thomson Reuters Institute’s 25th Annual Law Firm COO & CFO Forum

By Mike Raposa, Chief Revenue Officer, Williams Lea

Big Law is breaking profit records again. But for how long?

The 2026 Am Law 100 grew revenue 13% and profits per equity partner 14%, and firms are investing behind that growth. The Thomson Reuters Institute’s Law Firm Financial Index put technology and knowledge management spending up 11.6% in the second quarter of 2026, with overall overhead up 7.7%.

Look beneath the headline numbers and the strain is showing. Wells Fargo’s review of more than 140 firms found revenue was up 12.4% in the first half of the year, while inventory grew 17.7% and collections slowed by 5%.

Firms are spending more to run the business, carrying more unbilled work and waiting longer to get paid. AI is supposed to help with all three. So far, the returns are hard to find. McKinsey’s The State of AI in 2026, published in August, reports that eight in ten respondents say AI has improved their productivity, while 37% attribute an EBIT impact to it. And in the Thomson Reuters Future of Professionals Report 2026, 77% of clients call AI-enabled quality improvement from their firms very important or essential, while only 3% to 6% say that most or all of the firms they work with deliver it.

That gap is where the COO and CFO now sit. On Thursday, October 22, I will moderate The Convergence Era: Operations, Finance, IT and Risk in the Age of Legal AI at the Forum in Washington, DC. I will be joined by Cindy Thurston Bare, Chief Data & Innovation Officer at FBT Gibbons; Ann Rainhart, Chief Operating Officer at Fredrikson & Byron; and Karen Campbell, Chief Information Officer at Foley Hoag.

Last year at this same Forum, Laurel CEO Ryan Alshak said AI should “automate the laundry and the dishes, not the science and the arts.” A year on, the more useful question is who inside the firm is redesigning the laundry and the dishes. Here is what we will dig into.

Who owns the work that crosses departments

Law firms have always been organized by department. That works until the deliverable crosses all of them—and real AI workflows do. Every matter runs on work that moves across those lines: intake, matter opening, document processing, billing and production. Each department owns its piece of that chain, and the gains from AI stall at every handoff where ownership stops.

In the Blickstein Group’s 2026 Law Firm COO Survey of 213 COOs, the leading barriers were practice silos (38%), no strategic consensus (28%) and a lack of operational authority (27%). The org chart is converging, with new chief AI, data and innovation roles at firms across the market. The question is whether ownership of the work is converging with it.

Our panel brings three different answers. At Fredrikson & Byron, Ann oversees finance, IT, marketing, HR and practice management as COO. At Foley Hoag, Karen is the firm’s first CIO and sits on a business leadership team whose operations portfolio spans IT, finance and risk. At FBT Gibbons, Cindy leads the data and innovation team that ran the evaluation before the firm expanded Harvey to every lawyer and professional in August. I want to hear how each structure handles a workflow that touches all of them, and who owns it end to end.

Where AI is showing up in the numbers

About two-thirds of the COOs in the Blickstein survey do not formally measure AI efficiencies, and 18% of professionals in the Thomson Reuters report say their organization tracks AI ROI. Costs are rising at the same time. In McKinsey’s survey, one in five organizations reports limiting AI use because of token and operating costs, and the major legal AI vendors are moving toward pricing based on consumption.

The upside is large when firms look past individual tasks. At a recent COO and CFO roundtable in New York, one law firm leader estimated that moving 15 operational levers by 1% to 3% each could add $100 million to a large firm’s bottom line. Each of those levers is a process, and AI pays off when it moves several of them at once.

That is the pattern behind the gap. AI applied to an unchanged process gives you faster individuals and the same business. The returns come from redesigning the chain of work around it, and from deciding which workflows are worth their token bill.

IT as a partner in managing risk

Generative AI debuted at No. 2 on the list of security challenges in the ILTA 2026 Technology Survey, which covered more than 500 firms, and most firms now govern several AI platforms at once: Microsoft 365 Copilot is in 76% of firms, CoCounsel in 44% and Harvey in 43%.

The legal and governance stakes became more concrete this year, too. In February, Judge Jed Rakoff of the Southern District of New York held in United States v. Heppner that documents a defendant created with a public AI tool were not privileged, in part because the provider’s terms allowed it to retain and disclose user data. Where AI runs and whose terms govern the data are now part of the privilege analysis. In April, a leading New York firm told a bankruptcy court that its AI policies had not been followed on a motion that contained 42 errors. A policy is the starting point. A process that holds up on every matter is the goal.

Clients are raising the bar as well. Outside counsel guidelines now set terms for data handling, audit rights and AI use, and some firms are already using AI to keep up with them. Fredrikson & Byron uses generative AI to read client guidelines and bring billing restrictions to the surface at time entry. Meeting those terms takes IT, risk, finance and operations working from the same plan, with IT in the room as a strategic partner from the start.

People, structure and governance

Firms are creating hybrid roles that bridge IT, finance and risk. Bloomberg Law reported in July that Director of AI roles now pay up to $440,000, with at least 16 firms competing for them. Some of the best hybrid talent may already be in the building. As AI takes on more routine processing, the people who know that work can apply their judgment to a far greater volume of work.

How firms roll AI out matters as much as who they hire. Foley Hoag took Harvey firmwide in under a month, with mandatory ethics training from day one and a peer ambassador program, and reached 94% lawyer adoption.

Firms are also building centers of excellence that combine AI, knowledge management, training and pricing. Two lessons keep coming up in my conversations with firm leaders. Governance for the practice of law and governance for the business of law are different problems, and firms that run them as one tend to slow both. And most firms now have an AI policy, but the harder work is deciding who holds the authority to approve, prioritize and stop a project.

Every firm also has to decide how it will run its operational layer: build the capability in-house, standardize on one vendor, or bring in an embedded partner. Most will combine all three. Maturity means deciding who is accountable for the result.

Where to start

We will close with the one step each panelist would take in the next 90 days. Mine is the one I keep coming back to: map one process, simplify it, connect the systems, then choose the tool.

The Convergence Era runs Thursday, October 22, from 1:30 to 2:20 pm at the Conrad Washington, DC. If you are at the Forum and want to compare notes, https://hubs.la/Q04zGk830.