There's a common (alarmist) narrative of recent years: artificial intelligence is about to start massively taking jobs away from people. It's discussed at conferences, written about in articles, argued about in comments. But this story has a side that's mentioned much less often: AI is capable not only of displacing familiar professions, but also of creating new ones. Sometimes this happens so rapidly that the market doesn't have time to prepare the required number of specialists. Employers on both sides of the ocean are urgently looking for people who can simultaneously understand technology, legislation, and corporate risks — and are willing to pay quite attractive money for such expertise!
If I were asked to recommend a promising field of study to a high school student, I would definitely add one very young specialty to the list — AI compliance.
Where the AI compliance officer profession came from
On August 2, 2026, the EU AI Act — the European law on artificial intelligence — came into full force. Setting aside the convoluted legal phrasing running to three hundred words, the gist is fairly simple: companies operating in the European market and using «high-risk» AI systems must now prove to the regulator that their technologies meet the established requirements, that the neural networks aren't running wild. Also it doesn't matter where the company itself is registered — the law is the same for everyone. Documentation, risk assessment, audits, and compliance reports — the full bureaucratic package, without which using certain AI systems becomes problematic (if possible at all).
This is where an unexpected problem surfaced. Checking artificial intelligence for compliance requires a specialist who simultaneously understands several completely different fields: they need to grasp how machine learning works, be able to read complex legal documents — and at the same time explain the results of the check to management in plain, simple terms. Such a specialist is able to talk to a developer, a lawyer, and a director all at once. Such people are practically nonexistent, since as recently as five years ago this profession simply didn't exist as a separate field.
The shortage of specialists is already showing up in salaries
The scale of the problem is clearly visible in the numbers.
- According to LinkedIn Skills on the Rise data, AI governance job postings grew by 150% over the year: more than 14,000 positions are open. Individual roles, including AI Compliance Officer and AI Ethics Consultant, are also showing notable momentum — their numbers are increasing by roughly 45% year over year.
- Even more telling are the results of a survey by the IAPP — the world's largest organization of privacy professionals: 77% of organizations are already building AI governance programs, but only 1.5% are satisfied with how well-staffed their own teams are. In other words, almost every employer that has already taken up AI governance is running into a shortage of qualified personnel.
- According to Forrester analysts' estimates, by the end of 2026, 60% of Fortune 100 companies will have set up AI governance director positions.
The market is responding to the shortage with money.
- A typical salary range is $120–180 thousand a year, while senior specialists at big tech companies earn more than $200 thousand. At the same time, moving from classic compliance into the AI track can bring an extra 25–35% — the market pays a premium for skills tied to artificial intelligence and the EU AI Act.
- One of the most in-demand specializations is considered to be the AI auditor: their income is estimated at $130–188 thousand.
Within just a few years, a whole layer of new positions has emerged, but the market still doesn't have enough people capable of filling them.
What does an AI compliance officer actually do?
Picture a meeting where three specialists come together at once. The engineer says: «The model shows 94% accuracy, everything works as it should, I have no issues on my end». The lawyer objects: «But the legal requirements call for certain conditions regarding the training data, and we can't ignore them». The manager listens to both and asks the question that matters most for the business: «So will the company run into trouble or not? Is this definitely all legal?»
The AI compliance officer turns out to be exactly the person able to understand everyone in the conversation and translate their arguments from one professional language into another, bringing them to a consensus. They check what data the model was trained on, identify potential weak points in the system, record possible errors and risks, compare how the technology works against the regulator's requirements, and prepare the necessary documents, signing off on them personally. Essentially, this is a new generation of technical auditor. The logic is roughly the same as for a financial auditor, except that instead of accounting documents, the object of review becomes neural networks, the processes behind their use, and the risks tied to them.
The work may not seem like the most exciting or dynamic, but it has several serious advantages:
- It can't be fully handed off to ChatGPT: checking AI's compliance with legal requirements has to be done by a human.
- The profession doesn't depend as heavily on the latest wave of tech hype: the more AI systems there are and the more complex their regulation becomes, the more work there is for compliance specialists.
Why can't universities quickly train these specialists?
At first glance it seems logical enough: if the market is short on talent, universities should start studying such specialists en masse! However, AI compliance is a hybrid profession, while education around the world is still organized by separate tracks, by «guilds». A law school doesn't turn students into machine learning specialists, and computer science doesn't turn graduates into experts on EU legislation. That's why the people who hold such positions today often arrive at the profession by unconventional routes: among them you'll find lawyers who taught themselves Python, and data science specialists who additionally worked their way through the regulatory side.
The education system, as often happens, is responding to market shifts with a delay: specialized programs in AI law and tech policy are only just beginning to appear in European and American master's programs. Among professional certifications, the market currently recognizes AIGP — Artificial Intelligence Governance Professional from the IAPP.
The result is a classic situation: demand has already taken shape, while a full-fledged path for preparing specialists doesn't yet exist. Those who enter this field now may collect a kind of premium for the talent shortage. In another ten years, given how fast technology is developing, the picture will change: more dedicated educational programs will appear, the number of graduates will grow, but competition will grow right along with them. The combination of «law + data» or «computer science + regulation» already looks, in principle, like one of the most underrated educational paths today, and students should definitely take a look at it.
But there's an important geographic nuance: the profession got its major boost thanks to European legislation, so Europe is becoming the main market for AI compliance officers. Studying at a European university followed by an internship there could turn out to be one of the most direct paths into this profession (we've already written in detail about the fact that student visas for Russians haven't gone anywhere).
How long will this AI momentum last?
The new profession has an obvious weak point for skeptics to pick at. It arose on the back of a specific law, so the question comes up: what happens if the rules change or get relaxed? Won't AI compliance turn out to be just another short-lived bubble?
The objection is fair enough. True, there are no guarantees, but there's a historical precedent that allows you look at the situation differently. That's GDPR — the General Data Protection Regulation. When the EU adopted the personal data regulation in 2018, skeptics reasoned much the same way: this is just another bureaucratic fad that will gradually fade and stop being critically important. Today, eight years later, the role of DPO — Data Protection Officer, a data protection specialist — has become a standard position at thousands of companies worldwide, and the very idea of personal data regulation has spread far beyond the European Union: dozens of countries have created their own rules modeled on the European approach.
A similar picture is currently taking shape with AI regulation. Local rules are emerging in different parts of the world — from Brazil to South Korea. Even if you assume a moderately unfavorable scenario and imagine that certain provisions of the EU AI Act get revised under pressure from lobbyists, the skill itself won't disappear — the ability to understand both technology and law at the same time is far broader than any one specific law, and it definitely won't leave you without a livelihood. The bet here isn't placed on a single job title or a single piece of legislation — it's about an interdisciplinary competence that will stay in demand as long as technology keeps making its way into business and governments keep trying to set rules for it. As you can imagine, technology will only keep moving forward — which means the value of these new specialists will only keep growing.
The main takeaway
While everyone is busy arguing over whether artificial intelligence will take jobs away from programmers, designers, and other specialists, right next door another market is taking shape — professions that are emerging precisely because of AI's spread. AI compliance is one of the first notable examples of this process: the field has already grown into serious salaries, while companies have run into an obvious shortage of suitable specialists. This is far from the last profession that artificial intelligence will create!
The practical takeaway is especially interesting for anyone currently choosing their education. The most promising and well-paid professions of the next decade may not sit inside one single field — not purely in technology, and not purely in the humanities either. More and more opportunities are arising precisely at the intersection of disciplines: law and data, computer science and regulation, technology and ethics — that's exactly where professions that didn't exist until recently are appearing today. Try to choose interdisciplinarity, look beyond your core profession, and you really won't go wrong!
