Why GC-Led Innovation Stalls: 5 Insights from the Leaders Living It

Legal teams know AI is here. The harder question is what to actually do about it.

That was the thread running through a recent panel at LegalTech Talk London, where Blee CEO Guy Shahar joined Monica Rissom (Group General Counsel, Starling Bank), Rachel Gonzales (former CLO, GE Vernova and former GC, Starbucks), and Sapan Gupta (Group General Counsel, ArcelorMittal) to discuss what's really blocking technology adoption inside legal functions—and what it takes to move past it.

While you may have missed the live discussion, we captured the most important insights for you below:

1. The Question Is No Longer Whether the Technology Works

The panel opened with broad agreement that the inflection point has passed. As Guy put it, the conversation has shifted: it's no longer about whether AI can do something, it's about whether the organization can absorb it. The technology is moving faster than most teams can think through the change management implications—and that gap is where adoption stalls.

2. Orphaned Ownership Kills Momentum

One of the sharpest observations came early. Guy described a pattern Blee sees repeatedly across large enterprises: "orphaned ownership." Lots of people are excited about innovation. Far fewer are willing to lead it. Without a clear internal champion—someone accountable for driving adoption, not just endorsing it—projects lose momentum before they get traction. Finding those drivers, not just enthusiasts, is one of the first practical problems any legal team needs to solve.

3. The Resistance Isn't Always Where You Expect It

Rachel Gonzales noted that the feelings that tend to slow adoption—uncertainty, fear of job loss, anxiety about what efficiency gains will mean for headcount—are understandable and largely predictable. What surprised her more was the other end of the spectrum: teams that took matters into their own hands. At GE Vernova, a legal team within a software business built their own AI tool using internal engineering resources, and that tool ended up being deployed more broadly than the third-party options being evaluated at the same time. Motivation, when it exists, finds a way.

Sapan Gupta added a structural observation: the first obstacle often isn't the legal team at all—it's IT. The default response from internal technology teams is "we can build that." In a company the size of ArcelorMittal, legal simply isn't at the top of the build priority queue. His solution was to hire a dedicated legal tech resource—a non-lawyer who specializes in legal ops—and that hire accelerated adoption more than anything else.

4. The Business Case Has to Land Before the Budget Does

Monica Rissom described the moment her CFO got on board: not when she made the case for the tool, but when she demonstrated that it would free her team to focus on what they actually add value to—judgment. The tool handles the low-value tasks. The lawyers handle the decisions that require experience. Once that framing clicked, the savings target followed.

This framing—legal as an enabler of business growth, not a cost center—is one Guy reinforced from the vendor side. The teams getting the most from compliance and legal tooling are the ones that have made the shift from gatekeeper to partner. That shift requires leadership to articulate a clear problem first. As Rachel put it: people need to feel something before they'll get on board. Without a problem they recognize and a vision they can rally behind, the technology stays in pilot.

5. Pilot Purgatory Is Real, and the Market Makes It Worse

Sapan was direct about another friction point: the POC cycle that never quite ends. Every time a pilot approaches conclusion, someone surfaces a new test case—an untested jurisdiction, an unconsidered scenario. Meanwhile, the volume of legal tech vendors competing for attention is making it harder to decide, not easier. Constant inbound from competing tools keeps teams in evaluation mode instead of implementation mode.

The Judgment Question Doesn't Have an Easy Answer

The most unresolved conversation of the session was around what accelerated AI adoption means for how lawyers develop—particularly junior ones. Monica raised it plainly: her generation learned by reading documents, marking them up, sitting with the complexity until judgment formed through repetition. If AI shortens that process, do lawyers reach senior roles without the underlying experience that made the judgment meaningful? The magic circle firms she's spoken with are trying to find a balance and haven't cracked it yet. This isn't a reason to slow adoption. It is a reason to be deliberate about how training evolves alongside the tools.

Guy closed with a framing that captures where the industry sits: we're still so early that standards from six months ago no longer apply. The analogy that stuck was electricity in its first months, not a comparison to overstate, but a useful reminder that most of what will matter about this technology hasn't been built yet. Those who can absorb the pace of change and act on it, rather than waiting for it to stabilize, will be in the strongest position.

Want to watch the full session? Catch the recording here.

‍