Professional services face an AI reckoning from inside and out.

Professional service firms are embracing AI at breakneck speed – but are they capturing real value? A new global survey indicates that 40% of professionals in consulting, law, accounting and other advisory fields say their organizations now use generative AI, nearly double the 22% from a year ago ([1]). Over 80% of those users engage with AI tools at least weekly and expect them to become central to workflows within five years ([2]). Yet this enthusiasm comes with a dose of anxiety: while 66% of professionals are optimistic about GenAI’s future in their daily work, many also worry that AI could threaten jobs, traditional billing arrangements, and even the definition of professional roles ([3]).

Despite the rapid uptake, clear measures of success are lagging. Only 18% of professionals report that their organizations track the return on investment (ROI) from AI initiatives, and 40% don’t know if any ROI is measured at all ([4]). In practice, this means a wave of AI adoption is happening on faith – or fear of missing out – rather than proven financial benefit. This lack of metrics is leaving firm leaders in the dark about whether AI is truly boosting productivity or just adding cost. As one industry report notes, a disconnect is growing: Two-thirds of corporate clients want their outside firms to use AI, but fewer than 20% explicitly require it, creating confusion and mixed signals for service providers ([5]).

The result is a brewing challenge to the century-old business models of these firms. Many consulting and advisory practices still derive revenue from billable hours and large junior teams performing manual analysis – exactly the model that ubiquitous AI automation stands to erode. If AI can do in seconds what used to take a team of associates or analysts days, the traditional lever of “more hours equals more fees” stops making sense. Forward-looking analysts have warned that if an outside adviser can’t deliver measurable outcomes at the speed of AI, they risk becoming obsolete. Professional services executives are starting to acknowledge that AI’s economic impact might require new pricing models (such as outcome-based fees) and new definitions of productivity and value, rather than complacently relying on headcount and hours. The latest data underscores that while AI is here to stay, the firms who thrive will be those who reconcile high-tech efficiency with a compelling value proposition beyond brute-force effort.

No sector of professional services feels AI’s upheaval more acutely this week than the legal industry. In California, lawmakers just sent a clear signal that AI’s wild west days for lawyers are ending. The state’s legislature unanimously approved the nation’s first law governing how legal practitioners use generative AI, with Governor Gavin Newsom’s signature expected by the month’s end ([1]) ([2]). The bill prohibits attorneys from effectively delegating legal work to AI and bars inputting confidential client data into public AI systems ([3]). Crucially, it requires lawyers to take reasonable steps to verify the accuracy of AI-generated material – checking all case citations and correcting any “hallucinated” facts – and to disclose when AI has been used in preparing court filings ([4]). While the new law doesn’t prescribe specific penalties, violations could trigger existing sanctions and disciplinary action ([5]). This move – the first of its kind in the U.S. – embeds AI accountability into the legal code, and other states are watching closely ([6]).

At the same time, technology firms are racing to embed AI deeper into legal practice. In the last 48 hours, OpenAI rolled out “Astra for Law,” a specialized version of its latest GPT-6 AI model geared for legal research and drafting. The system is enhanced with a massive index of over 230 million legal documents (cases, statutes, regulations) and tailored instructions for legal analysis ([7]). A handful of large law firms have been granted early access via a secure program that offers features like “Trusted Access” and zero data retention, addressing confidentiality worries in using AI tools ([8]). OpenAI’s legal AI has already demonstrated significantly higher accuracy on legal research questions (54% correct answers) compared to the base model’s performance (38.7%) ([9]). For law firms and in-house counsel, this isn’t just another tech demo – it’s a live product that forces decisions on how to integrate AI into real legal workflows ([10]).

The confluence of stricter oversight and cutting-edge AI capabilities is pushing law firms toward a new equilibrium. On one hand, judges and regulators are clamping down on careless AI use: more than 500 court orders in the U.S. have already addressed the problem of fake, AI-invented case citations, with some courts even issuing standing rules on AI use in filings ([11]). On the other hand, lawyers now have access to AI tools powerful enough to perform tasks once reserved for junior associates or research staff – from analyzing contracts to drafting briefs – in a fraction of the time. The message is that AI in legal services is moving from experiment to infrastructure ([12]). Firms must simultaneously harness these tools to stay competitive and vigilantly govern their use to avoid high-profile missteps. The California bill and OpenAI’s new platform, arriving virtually together, highlight both the promise and the perils of this transition.

It’s not just internal transformation or regulation pressuring traditional firms – it’s also emerging competition and empowered clients. In-house teams are rapidly adopting AI solutions that reduce their dependence on outside advisors for routine work. A poignant example is this week’s launch of Lightfield, a Y Combinator-backed legal AI startup that acts as an autonomous corporate paralegal ([1]). Lightfield’s platform plugs into a company’s own communications and document systems to gather context and draft legal documents before the human legal team even gets involved ([2]). By handling intake, research, and first drafts of contracts or compliance paperwork automatically, such “agent” systems promise to free up in-house counsel – and cut down on basic tasks that might previously have been sent out to law firms for a fee ([3]).

These kinds of AI-driven alternatives are proliferating across professional services. In the legal arena, we are even seeing the rise of “AI-native” law firms that claim to deliver outcomes faster and cheaper by heavily automating research and document production. And in the consulting world, corporate clients are deploying advanced analytics and GPT-powered tools to analyze their own data and generate strategic insights, reducing the need to hire external consultants for tasks like financial modeling or routine analysis. This trend is feeding a fundamental question for the industry: if clients and competitors have access to the same powerful AI tools, what unique value can traditional firms provide? The competitive moat of owning superior knowledge or analytical capacity is narrowing. It’s a wake-up call that the next generation of professional services may be built as much around technological prowess as around headcount or legacy reputations.

In an AI-saturated services market, the role of human experts is rapidly evolving rather than disappearing. Intelligent systems may handle research, number-crunching, and first-draft prep, but they still lack the seasoned judgment, empathy, and ethical nuance of experienced professionals. Even OpenAI’s new Astra platform stresses that final responsibility for accuracy and judgment remains with the lawyer using it ([1]). Likewise, a key element of training junior lawyers now involves learning how to critically assess AI outputs and knowing when human intervention is needed – a skill one major firm has deemed so essential that it recently made a rotation in its “AI Knowledge” department mandatory for all new trainees ([2]).

To maintain an edge, professional firms are beginning to invest in people who can bridge the gap between AI capabilities and business needs. This week, a large U.S. law firm appointed its first Director of AI Development to build custom AI applications for the practice ([3]) ([4]) – part of a broader trend of top firms establishing dedicated AI leadership roles to drive strategy and governance ([5]). Across consulting and accounting firms, similar roles and training programs are emerging to ensure practitioners can leverage AI in ways that enhance (not replace) their expertise. The consensus is that much of the "grunt work" in research, data analysis, and routine document drafting will be increasingly automated, shifting human professionals into a higher-value role.

That high-value role is where “expert human judgment” must be redefined and emphasized. In practice, this means doubling down on the insight, creativity, and strategic perspective that machines cannot replicate. The Thomson Reuters Institute report accentuates that AI’s value is unlocked only by effective strategy, execution, and infrastructure – areas that depend on human leadership ([6]). For managing partners and C-level executives, the mandate is clear: success in the AI era will rest on combining powerful AI tools with the irreplaceable human capacity for context, leadership, trust-building, and ethical judgment. Professional services firms that can integrate AI for efficiency while clearly demonstrating human insight and accountability will be best positioned to thrive in an environment where technology alone is never a substitute for true expertise.