Writing from New York, where I am attending (and co-organizing) a wonderful UIA event on wellness in the legal profession. As tends to happen on these trips, the formal program is only half of it. The other half is the networking around it: meetings with law firms, conversations with corporate legal departments, catch-ups with friends, dinners, and endless coffees.
Yesterday I sat down with a partner at one of the largest Italian firms. We talked about wellness for ten minutes and about artificial intelligence for an hour and a half.
That AI is going to reshape the legal market is no longer up for debate. The interesting questions live elsewhere: how, how fast, and on whose terms. Will we see AI-native law firms? Will private equity finally get a real way in? Will the Big4 continue their rise? Will the biggest service vendors quietly pivot and become law firms themselves?
These are good questions. What genuinely fascinates me, though, sits one layer further down, in the regulatory layer.
If you read the codes of conduct for lawyers in most jurisdictions, as well as the laws regulating the legal profession, you are reading a document written for a profession that, in most cases, no longer fully exists. They reflect a very old way of being a lawyer: solo or in small partnerships, no external capital, no advertising worth mentioning, a strong identification between the person and the practice. They also read as if litigation were the only legitimate way to be a lawyer, even though litigation accounts for around thirty percent of the Italian market. Italy is more conservative than most, but the discussion is alive almost everywhere, and especially in civil law countries. It just moves slower than the technology.
I know this first hand.
When I launched Better Ipsum, I wanted to build a benefit corporation. Fully remote. Working at the intersection of AI, legal design, and legal operations. Serving both law firms and corporate legal departments. Running real use cases. Open to external capital when the moment was right. And I wanted to be the one running it.
The answer from the bar was straightforward: you have to resign.
So I resigned.
From a strictly financial point of view, it was not the wisest decision I have ever made. The social security regime for Italian lawyers is, on paper, less generous than some of the alternatives, and you give up a few defaults that are convenient when you operate inside the profession. I did it anyway, because the alternative was building a smaller, more compromised version of the company I actually wanted to build.
The moment you step outside, you start to see the cost of staying inside very clearly. Imagine having a law firm with revenues gravitating around 100 million euros per year.
Want to open up to private equity? You can’t.
Want to list on a stock exchange? You can’t.
And it does not get easier as you scale down.
Want to publish a track record? You can’t, not in the way other professionals can.
Want to publish detailed case studies? Same answer.
Want to advertise? Yes, but “cum grano salis”, with so many caveats that meaningful marketing in the legal field becomes a small craft of its own.
The fun part is that the more we cling to the law firm as an “association of professionals,” the less we end up investing in people and technology. Think about it. If things go badly, say a war or a pandemic or a sharp downturn, the partnership has no real buffer to keep its people on the payroll. On the technology side, only a corporate structure makes serious multi-year investment plausible. Partnerships built around annual distribution are structurally allergic to long-horizon spend.
And I could keep going. The market would benefit enormously from being able to read the financial statements of law firms, from associates being recognised as the de facto employees most of them already are, from external auditors being allowed to look inside accounts and practices the way they do in any other industry. There are small experiments here and there. But they are too small to be worth listing.
The largest Italian firms have, of course, found ways to live with this. Some run parallel corporations for the work that does not fit inside the professional shell. Some have, in practice, lawyers de facto administering structures that the rules say they should not be administering. All of it is the predictable consequence of asking modern professional service businesses to operate inside a nineteenth-century regulatory cage.
I care about this for personal reasons. I was the first head of innovation in a law firm in my country. I spent most of my professional life thinking about the future of legal services. I wrote about proactive law, gamification, and human-centered AI before it was a cool thing. But more than anything, I care because I believe clients deserve more than the current settlement allows.
Take litigation funding. In my former firm, I opened the conversation with the funds together with two partners of the litigation department. We were among the first (if not the first) to do so in Italy. The model is, on its face, beautiful. The firm gets paid. The client does not pay out of pocket. The fund takes a slice if the case wins, and absorbs the loss if it does not. Everyone’s incentives line up around the merits. And yet the conversation around it in Italy has been, for years, an endless negotiation with bar associations and an underlying fear that one regulator, on one bad morning, will decide that the whole arrangement is incompatible with professional independence.
That fear has precedent. Look at what happened in France in 2019. The regulator decided that you could not run predictive analytics on judicial decisions, at least not in the ways the legaltech companies had been doing. From one day to the next, an entire generation of products built around judge analytics had to rethink their offering, and several of them simply stopped sending their alerts. Years of product work, gone, because the regulatory frame moved.
The mirror image is also possible. Yesterday I scrolled past a post on LinkedIn from a former colleague who is launching something I genuinely like: you generate your own contract or template with AI, and a lawyer reviews and validates it before you use it. Essentially the legal version of asking AI for a diagnosis and a doctor for the prescription. Will it work? Probably yes, in some configurations. Will it be without regulatory friction? Almost certainly not.
In the meantime, I look at the UK and I see structures like Garfield Law, an AI-driven legal service that received approval from the Solicitors Regulation Authority. The conversation there has already moved past “can we do this?”, and is now about what advertising and transparency should look like. That is the gap in a single image. In one country, the question is whether the technology should be allowed to exist. In another, the question is how it should be marketed.
Meanwhile, the capital is moving regardless. Andreessen Horowitz, Sequoia, and others are placing serious bets on AI-driven services companies, including in legal. The market will respond, because markets always do. Demand from clients is real, margins and multiples are genuinely appealing, the technology is improving every quarter, and the unit economics are starting to make sense. What I am less sure about is whether the regulatory environment will move at anything like the same speed.
Here is the part I keep coming back to.
The entire modern technology industry has been built on a single operating principle: move fast and break things. Amazon broke retail. Uber broke private transport. Netflix broke television. YouTube broke a great deal at once. OpenAI and Anthropic are reshaping how knowledge work gets done, with consequences we are only beginning to understand. None of these companies were welcomed by their regulators. All of them changed the rules by force of adoption. Paradoxically, I am not a fan of this trend. As a lawyer who also teaches copyright and fashion law, I carry my own complicated relationship with that operating principle, more than people might guess.
However, one thing is certain, at least to me. The legal profession will not be excused from this paradigm. It will probably resist longer, more eloquently, and with better footnotes than any other industry. That is what a profession built on precedent does.
That said, the question I am carrying out of this line of thought has little to do with whether the bar will eventually update its rules. It will. What I keep weighing is how many good ideas, how many useful companies, and how many better outcomes for clients we will lose in the meantime.
I resigned from the bar, hopefully only temporarily, to find out. Italian rules leave the door open: I can be readmitted later. And I can still call myself a lawyer, even if not a practicing one.
And no, I do not regret resigning.
But I would prefer, for everyone coming behind me, that it not be the entry ticket.
