Professional Services Face AI’s Double Disruption

Artificial intelligence is steadily eroding the foundation of the traditional talent pyramid in consulting, law, and accounting. New data shows entry-level hiring is already contracting sharply. In the UK, graduate-level job vacancies have plunged by 34.9% in one year ([1]), with analysts pointing to AI-driven automation of research, drafting, basic analysis, and other junior tasks as a key factor. A global HR survey found businesses now entrust roughly one-third of tasks once handled by fresh graduates to technology, rising to nearly 40% in markets like the UK ([2]). In some organizations, AI performs over half of all entry-level work ([3]) – from due diligence document review to initial financial modeling – upending the workloads that have long kept armies of junior staff busy.

Professional firms are responding by weaving AI into their processes, aiming to boost efficiency while redefining roles. For example, law firm Shoosmiths recently launched a proprietary contract review AI built with Microsoft’s Azure OpenAI, promising to cut “several hours” of human work per document and standardize advice across deals ([4]). They frame the multi-million-pound initiative as a way to scale their mergers-and-acquisitions practice, following peers like Allen & Overy (partnered with legal AI startup Harvey) and Clifford Chance (piloting tools like Kira for contract analysis) in investing heavily to automate routine work ([5]). In consulting, internal AI assistants now help consultants handle research, slide generation, note-taking, and even writing in the firm’s signature style ([6]). McKinsey & Company, for instance, envisions a future 1:1 ratio of human consultants to AI “colleagues,” using bots to perform many tasks that demand only “mediocre expertise” while humans focus on higher-value activities requiring judgment ([7]).

These trends raise an uncomfortable question: How will firms develop future experts if machines do the entry-level work? The answer may lie in reinventing junior roles rather than eliminating them. In one study, a remarkable 96% of HR leaders said they expect today’s entry-level jobs to evolve into “AI supervision” positions within five years ([8]). Instead of churning through spreadsheets or case law, tomorrow’s new hires will likely spend more time validating AI outputs, training machine learning models with domain knowledge, and tackling nuanced client problems that algorithms can’t easily solve. In the meantime, however, the traditional apprenticeship model – where junior staff learn by doing the grunt work – is under strain. Firms face a potential experience gap if they cannot find new ways for the next generation to develop the judgment and contextual expertise that only comes with hands-on practice.

The flipside of the AI disruption story is unfolding on the client side. Just as service providers adopt clever tools, their customers are doing the same – and becoming less dependent on outside help in the process. Now that generative AI puts unprecedented analytical power directly into the hands of executives, companies can analyze their own data, generate insights, and test ideas without hiring outside teams ([1]). Corporate legal departments, for example, more than doubled their use of AI in the past year (from 23% to 52%), far outpacing the adoption rate at many law firms ([2]). In fact, 64% of in-house legal teams now expect to depend less on external counsel thanks to AI tools they’re building internally ([3]). This trend threatens a core revenue stream for major law firms – routine work from corporate clients – unless those firms can offer capabilities that clients don’t have.

The story is similar in consulting and accounting. After years of paying hefty fees for large teams to crunch data and deliver reports, many enterprises now see AI as a way to get faster results with fewer outside consultants. As one industry observer put it, the era when executives were willing to tolerate “bloated teams and formulaic processes” is ending ([4]). Management consultancies have already felt the shift, with reports of slowed growth and layoffs in top firms ([5]). The message from clients is increasingly clear: if an advisor’s value rests mainly on manual analysis or templated reports, AI may beat them on speed and cost. This is pushing buyers to re-evaluate when to bring in external experts. They’re reserving traditional firms for high-stakes, strategic judgments – and even those engagements come with higher expectations on efficiency. Day-long research projects and armies of junior analysts are a much tougher sell when an AI model can parse data or benchmark a market in seconds.

In this new reality, clients’ expectations of their service providers are rising. Many corporate leaders now demand that outside advisors bring not just expertise, but also proprietary AI tools and proven outcomes to the table. There’s also a growing call for transparency: in one recent survey, 60% of in-house counsel weren’t sure if their law firms were even using generative AI in their work ([6]). That gap will not be tenable for long. Clients want to know their advisors are leveraging the best technology available – securely and ethically – or else they may take those tasks in-house. The balance of power in professional services is shifting toward the client, as AI lowers the barriers to self-service problem solving.

It’s not just clients doing more for themselves – an influx of new competitors, from tech behemoths to well-funded startups, is rapidly changing the professional services landscape. In the last 48 hours, Amazon’s cloud division (AWS) unveiled a $1 billion "Forward Deployed" AI engineering program to embed its experts directly inside client companies and build advanced AI systems on the ground ([1]). This move by a Big Tech firm into hands-on implementation services puts it in direct competition with IT consultancies and the Big Four, which have traditionally been the go-to for enterprise technology projects. It follows a playbook pioneered by the likes of OpenAI, which last year launched its own high-touch consulting arm targeting $10+ million enterprise deals and embedding engineers within client teams to deliver AI transformations ([2]). Tech companies are effectively saying to clients: we can provide the software *and* help you use it – potentially cutting out the middleman consulting firm.

Meanwhile, a wave of AI-native professional service platforms is rising. Legal AI startup Harvey, for instance, recently secured $200 million in funding to reach an $11 billion valuation ([3]) – a level that rivals the market capitalization of many established law firms. Harvey’s product acts as a generative AI co-pilot for lawyers, already serving major global law firms and corporations ([4]). In another example that sent shockwaves through the UK legal sector, an AI-driven platform called Lawhive acquired a traditional law firm outright – the first deal of its kind ([5]). Lawhive’s virtual legal assistant, "Lawrence," can draft contracts, perform case research, and handle routine tasks normally assigned to paralegals ([6]). By fusing technology with a regulated law practice, these newcomers aim to deliver services faster and cheaper than the old partnership model. Professional services incumbents now face competition on multiple fronts: not only must they compete with each other, they must also contend with software vendors morphing into service providers and startups reimagining what an advisory firm looks like.

As AI commoditizes many tasks, the economics of advice are under pressure. Much of the traditional consulting and legal industry has been built on the leverage model – a pyramid of billable hours billed by junior staff and reviewed by seniors. But if AI can complete a week’s worth of analysis in minutes, that pyramid starts to crumble. A new report from HFS Research finds that 65% of enterprises believe the old model often fails to deliver tangible value, and 83% say AI-powered, outcome-focused consulting delivers superior results ([1]). “Headcount-based” fee structures are on the way out: today about half of consulting contracts pay by the number of staff or hours, but only 16% of clients expect to keep using that model within two years ([2]). In response, firms are exploring new approaches like value-based pricing, subscriptions, and success fees where they share more risk and reward. The Big Four are already shifting toward solutions and managed services that blend technology with ongoing support. PwC, for example, has publicly set a goal for 20–25% of its global advisory revenue to come from multi-year managed service contracts leveraging AI and automation ([3]). These deals emphasize continuous outcomes and technology platforms over one-off reports, offering clients more efficient results while challenging the hefty margins of old project-based billing.

This transformation poses a fundamental question for incumbents: what exactly are clients paying for when so much “work” is done by software? Advisory firms will need to convincingly answer that with new value propositions. Some are investing massively in proprietary technology to differentiate – Kirkland & Ellis, the world’s highest-grossing law firm, just committed an unprecedented $500 million to build an internal AI platform ([4]). But tools alone won’t justify premium fees. Ultimately, professional firms may have to provide guarantees of impact (for example, tying fees to measurable results) and focus on uniquely human strengths like contextual insight, creativity, ethical judgment, and the ability to integrate solutions into real-world organizational change. As one industry research leader bluntly put it, “Consulting as we’ve known it is over… If your consulting partner can’t deliver measurable outcomes at the speed of AI, they’re obsolete” ([5]). In short, the days of billing large teams to simply analyze and advise are numbered – clients want partners who can drive change and stand behind the results.

The rise of AI is forcing a re-examination of what "expert human judgment" truly means in these fields. In many cases, human expertise is becoming less about labor-intensive number-crunching or document review and more about guidance, oversight, and strategy. Leading firms insist that humans remain at the core of their value. For instance, PwC’s new AI-enabled delivery model is explicitly built around keeping human judgment 'at the center of every critical decision' ([1]). Rather than replacing professionals, AI can serve as a force-multiplier – handling the heavy lifting of data processing while consultants, lawyers, and accountants interpret results, provide creative solutions, and ensure decisions align with broader business and ethical considerations.

However, maintaining trust in an AI-saturated advisory process will require visible human stewardship. Regulators are already signaling that accountability must stay with humans: the UK’s Solicitors Regulation Authority recently authorized its first AI-only law service, but only after ensuring that qualified solicitors would supervise all AI-driven work and be held responsible for any errors ([2]). This underscores that judgment, ethics, and client trust still rest on human shoulders, even as software automates much of the grunt work. Professional judgment in an AI world may therefore shift to managing and curating AI outputs – making sense of what the algorithms produce and tailoring it to each client’s unique context. This is a very different role for the expert advisor than in decades past.

For leaders of professional service firms, the challenge now is to champion this new kind of expertise. It means retraining and empowering their workforce to work alongside AI, and being frank with clients about how human insight adds value beyond the machine. The firms that thrive will be those who leverage AI to amplify – not replace – their experts, and who can articulate clearly why that expertise still matters. In a world where clients have more tools and alternatives than ever, the human edge must be sharpened and redefined, not assumed.