Ad/Marketing Communication
This legal article/report forms part of my ongoing legal commentary on the use of artificial intelligence within the justice system. It supports my work in teaching, lecturing, and writing about AI and the law and is published to promote my practice. Not legal advice. Not Direct/Public Access. All instructions via clerks at Doughty Street Chambers. This legal article concerns AI Law
Introduction
Many of the most interesting conversations I have about AI take place after speaking engagements, or privately with other lawyers at parties or during entirely unrelated litigation. They often begin with, “Oh, you are that AI guy!” I always hope that this is not a comment on how little I resemble my chambers profile picture, but you never quite know…
What it does suggest, however, is that my aim of sharing as much practical and legal thinking about AI with the profession as I can is starting to take hold. I am genuinely grateful to the many people who share my work and to the readers and visitors to this blog and my other writing on these issues.
This piece is a short one but based on the conversations I have had over the past year, it feels timely to address a question that I am asked frequently, although often framed in different ways. Sometimes it is put politely: “Is AI law actually an area of law?” or “Are you receiving briefs in this area?” At other times, it is expressed more bluntly, ranging from “That snake oil law you practise” at one end, to “AI law will be the only area of law one day” at the other.
A defined speciality or something that cuts across everything?
I encountered more resistance than I expected when I have suggested, on a number of occasions, that AI law may become something lawyers actively practise as a defined speciality, rather than simply a convenient label to describe a potentially relevant issue permeating other areas of law. The tone of those responses has varied, but the underlying point has been consistent. If AI touches everything, how could it sensibly be its own area?
That instinct is an understandable one and I still recognise its force. If anything, the past year has only reinforced how rapidly AI is seeping into the day to day reality of almost every legal practice, if not more broadly into life more generally.
However, my view remains that AI law should be treated as a specialist area and in practice it already is becoming one.
I have never been comfortable describing AI as a tool. I need to do a more detailed post on why, but AI is now undeniably being used to make and shape decisions across multiple professions. How far it is being used and how far human decision making is being delegated to AI, is extremely important in many areas. This includes government functions, employment, law, medicine, housing, finance, and the wider public sector. The consequences of disputes arising specifically from the use of AI are now unavoidable.
I am already being instructed in matters involving the inappropriate use of AI, and we are likely to see some significant decisions emerging, particularly in public law. Those cases raise recurring issues that do not fit neatly within any single existing discipline. They involve questions about professional competence, disclosure and consent, record keeping, bias, explainability, evidence, and causation. They also raise difficult procedural questions, including what must be disclosed about AI use, how AI generated material should be treated, and what counts as proper expert evidence. The same legal problems are appearing in different settings, which is precisely why this work is beginning to cohere into a specialist area rather than remaining an afterthought within other fields.
Lexology and the market response
The market itself is beginning to respond to that reality. AI related work is increasingly being treated as a coherent category, with its own recurring risks, questions and disputes. One small but telling illustration of this is that the Lexology Index now includes a dedicated Artificial Intelligence report, within which there is a distinct contentious strand. Lexology’s own description of this report is instructive:
“Lexology Index: Artificial Intelligence identifies the leading lights and trailblazers in this rapidly evolving legal field, highlighting lawyers advising on AI-related matters across three areas: transactional and advisory work, regulatory compliance, and contentious issues. Recognised practitioners handle deals and governance, navigate GDPR and competition rules, and manage disputes involving IP, liability and insurance in the AI sector.”
So what is AI law?
I have come to think of AI law as two related, but not necessarily overlapping, things. The first is AI specific regulation and governance, where the legal rules are explicitly directed at AI systems or models. The European Union’s Artificial Intelligence Act is the most obvious example and it is now a lived reality, rolling out on a staged timetable.
The second is the much larger body of ordinary law that becomes newly complex, newly stretched, or newly uncertain when AI is involved. This affects too many areas to catalogue exhaustively, but it includes the kinds of issues I have already mentioned above such as failures to use AI where it may have been appropriate, the use of hallucinated authorities in legal work and the satellite litigation that can follow. It would also include procedural questions and issues around disclosure, evidence and transparency in AI assisted decision making.
These are distinct forms of legal work and there is no reason to assume that those practising in one will necessarily practise in the other, but each increasingly calls for lawyers who develop focused expertise rather than treating AI as a peripheral concern.
Conclusion
When I looked at my own Lexology profile recently, I saw that I am listed with an Index Ranking of “Recommended” in the category of Artificial Intelligence, and specifically Artificial Intelligence Contentious. I am ranked with 3 other Barristers, 246 lawyers, 133 firms across 39 countries. I am grateful for that recognition. Particularly because it reflects the direction of my work and writing.
However, what really matters is what comes next. If this many lawyers across so many jurisdictions are already being identified for AI related work, the obvious question is how quickly that number grows. Over the next year, will we see it double, or will it increase tenfold as disputes harden, regulation bites, and courts are asked to grapple with AI in earnest?
My instinct is that we are still at the very early edge of something much larger. For lawyers who are curious, prepared to learn, and willing to engage seriously with how AI is reshaping legal problems, there is an opportunity here to help define a practice area rather than simply react to it once it has fully arrived.
So what do you think? Do you see AI law becoming a distinct practice area in its own right, in the way that insolvency or public law is distinct? Or do you expect it to permeate legal practice so thoroughly that the label eventually becomes redundant, simply part of what it means to be a modern lawyer? There is more analysis to follow shortly. If you find these articles helpful, please consider subscribing below.
Final Words from Chat GPT 5.2 Pro and Gemini 3 Pro
This is where I ask premium LLMs to comment on my article above. So what do you think? Do you see AI law becoming a distinct practice area in its own right, in the way that insolvency or public law is distinct? Or do you expect it to permeate legal practice so thoroughly that the label eventually becomes redundant, simply part of what it means to be a modern lawyer?
Chat GPT 5.2 Pro
“AI law will almost certainly become a distinct practice area, but not because it is sealed off from the rest of legal work; rather, because it creates a repeating cluster of issues (competence, disclosure/consent, record-keeping, bias, explainability, evidence, causation, and procedural questions about what must be revealed and how AI‑generated material should be treated) that recur across domains and therefore reward focused expertise, especially in contentious matters, while AI‑specific regulation and governance (for example, regimes explicitly aimed at AI systems) simultaneously creates a recognisable advisory/regulatory track. In practice that points to the “both/and” outcome hinted at on the page: a specialist bar and solicitor market that develops deep capability in AI‑driven disputes, governance, and compliance, alongside a rising baseline expectation that every modern lawyer can spot AI issues in their own field—much as data protection once felt niche but is now ambient. The label “AI law” might eventually narrow (or fragment into sub‑labels like AI governance, AI litigation, and AI assurance), yet the underlying work will remain identifiable enough to justify dedicated practitioners, training, and market categories”
Gemini 3 Pro
“We are witnessing a divergence rather than a merger: while the use of AI will inevitably permeate every practice area as a baseline skill, ‘AI Law’ is rapidly crystallizing as a distinct, vertical specialty. The distinction lies between operation and governance. Just as the internet’s ubiquity did not dissolve ‘Cyber Law’ but rather deepened the need for experts in data sovereignty and network liability, the ubiquity of AI is generating a specific, dense thicket of regulatory and liability issues—from algorithmic bias to hallucinations and IP disputes—that cannot be navigated by generalist principles alone. The future will demand lawyers who do not just use the machine, but who understand the specific legal architecture that prevents the machine from causing harm.”