Why the AI Consulting Gold Rush Turned Into Fools Gold

The consulting industry faces a credibility crisis

The consulting industry faces a credibility crisis. After investing billions of dollars and deploying massive marketing campaigns promising to lead corporations through the artificial intelligence revolution, firms from the Big Four to elite strategy houses are confronting an uncomfortable reality: their clients increasingly believe they have overpromised and underdelivered.

The Wall Street Journal recently reported that companies ranging from pharmaceutical giants to mortgage lenders are ending consulting engagements and bringing AI implementation in-house, citing a fundamental expertise gap. This development represents more than a temporary setback for advisory firms. It signals a potential inflection point in how enterprises approach transformative technology adoption and raises critical questions about the consulting industry's value proposition in an age of rapid technological change.

The Expertise Illusion

The core issue is elegantly summarized by Greg Meyers, chief digital and technology officer at Bristol-Myers Squibb, who noted that a partner at a Big Four firm has no more experience with cutting-edge AI tools than a college student experimenting with the technology. This observation cuts to the heart of consulting's traditional model: packaging and selling expertise derived from cross-industry experience and repeated implementations.

Generative AI breaks this model because true expertise simply does not exist at scale yet. The technology emerged into mainstream consciousness with ChatGPT's release in late 2022, giving no one sufficient time to develop the battle-tested playbooks that consulting firms traditionally sell. Unlike enterprise resource planning systems, cloud migrations, or other technologies where consultants could learn from hundreds of implementations and develop standardized approaches, AI remains fundamentally experimental.

This represents a departure from historical patterns. When enterprises adopted cloud computing or implemented SAP systems, consultants could legitimately claim superior knowledge. They had seen multiple deployments, understood common pitfalls, and developed methodologies to navigate complexity. The technology itself, while sophisticated, operated within relatively predictable parameters.

Generative AI is different. Its capabilities evolve monthly, its applications are highly context-dependent, and successful implementation requires deep domain knowledge that external advisors struggle to acquire. As Tilak Mandadi of CVS Health observed, healthcare's complexity means internal teams are better positioned to identify viable use cases than external consultants, regardless of their general AI knowledge.

The Proof-of-Concept Trap

Perhaps the most damaging critique comes from clients like Merck, whose CIO Dave Williams noted that consultants often succeed at building proofs of concept but fail to scale them across the business. This failure mode reveals consulting's structural limitations when confronting truly novel technology.

Building a demonstration that generative AI can perform a specific task is relatively straightforward. Scaling that demonstration into a production system that delivers consistent business value requires deep integration with existing processes, data infrastructure, and organizational capabilities. External consultants, no matter how skilled, lack the institutional knowledge and sustained engagement required for this work.

The consulting model traditionally involves relatively short-term engagements where firms parachute teams into client organizations, deliver recommendations or implementations, and move on. This approach works for bounded problems with clear solutions. It fails when the challenge involves ongoing experimentation, iteration, and organizational learning.

Magesh Sarma of AmeriSave Mortgage articulated this realization bluntly: consultants proved just as good or as bad as what the company could have done in-house. When clients reach this conclusion, they naturally question why they are paying premium rates for external help.

The Internal Capability Gap Has Closed

The consulting industry's AI struggles also reflect a broader shift in enterprise technical sophistication. Large corporations today employ thousands of software engineers, data scientists, and technical product managers. Their internal capabilities for technology implementation have grown substantially over the past two decades.

This evolution began with cloud computing, accelerated through digital transformation initiatives, and reached a new level with AI. Companies that once relied heavily on external vendors for basic technical work now operate sophisticated engineering organizations. CVS Health, Bristol-Myers Squibb, and Merck all maintain substantial internal technical teams capable of experimenting with and deploying new technologies.

Michael Mische, a former KPMG principal now teaching at USC, argues that consulting firms were too slow to hire people with genuine AI competence, leaving them in a position of great vulnerability. But the deeper issue may be that no amount of hiring can overcome the structural advantages that internal teams possess: intimate knowledge of business processes, direct access to proprietary data, and the ability to iterate continuously rather than working within the constraints of a consulting engagement.

This shift mirrors patterns in software development, where enterprises increasingly prefer to build capabilities in-house rather than outsource to systems integrators. The consulting industry must confront the possibility that for certain types of work, particularly involving cutting-edge technology requiring deep integration and continuous learning, the traditional external advisory model simply does not fit.

The Revenue Reality Check

Despite public statements about growing pipelines and increasing demand, the financial evidence suggests consulting firms are not capturing the AI opportunity they anticipated. Global spending on generative AI consulting reached $3.75 billion in 2024, up from $1.34 billion in 2023, according to Gartner estimates. While this represents growth, it pales in comparison to the hundreds of billions corporations are investing in AI overall.

Accenture, the only major consultancy that is publicly traded, reported a $100 million increase in new generative AI bookings in its most recent quarter, down from $200 million quarter-over-quarter increases in the previous two periods. This deceleration suggests that initial enthusiasm for consulting support may be waning as clients experience the expertise gap firsthand.

Pat Petitti, CEO of Catalant, a platform for freelance consultants, reports hearing repeatedly from executives who paid consultants $20 million and received lengthy reports about AI's future without any practical application. This feedback indicates that some firms have fallen back on their traditional strength of producing impressive strategy documents while failing to deliver implementation value.

The economic pressure on consulting firms extends beyond AI. The industry has faced macroeconomic headwinds and conducted layoffs, making the AI opportunity even more critical for future growth. Firms wagered that generative AI would produce a sustained boom in advisory work similar to previous technology waves. Instead, they are discovering that this technology revolution may bypass them.

The Second Wave Hypothesis

Fiona Czerniawska, CEO of Source Global Research, offers a more optimistic long-term perspective for consulting firms. She argues that consultants have mistakenly tried to position themselves at the cutting edge, which is not where they belong. Instead, their opportunity will come in a second wave, perhaps four to five years from now, when generative AI becomes more mature, reliable, and predictable.

This hypothesis aligns with historical patterns. Consulting firms typically thrive when deploying established technologies at scale across industries. They add value by bringing standardized methodologies, managing change, and helping organizations adopt proven practices. If generative AI evolves from experimental technology to established capability with clear best practices, consultants could indeed play a major role.

McKinsey Senior Partner Eric Kutcher continues to tell CEOs that effectively leveraging generative AI could double their share price within five years. This bold claim reflects consultants' ongoing belief in AI's transformative potential and their own future role in capturing it. However, Kutcher also acknowledges that few clients are taking advantage of AI's full potential today, implicitly conceding that the promised transformation has not yet materialized.

The second wave hypothesis raises important questions for both consultants and their clients. If it proves correct, enterprises that develop strong internal AI capabilities now will be positioned to deploy at scale when the technology matures, potentially reducing their dependence on external advisors. Conversely, companies that wait for consulting firms to develop proven playbooks may fall behind competitors who are learning through experimentation today.

Implications for Enterprise Strategy

The consulting industry's struggles with AI implementation offer several critical lessons for business leaders navigating their own AI strategies.

• Enterprises should be skeptical of external claims to AI expertise, regardless of the source. The technology is too new and too rapidly evolving for anyone to possess comprehensive mastery. Leaders who recognize this reality can make more informed decisions about where external help adds value and where internal development makes more sense.
• Successful AI deployment appears to require sustained organizational learning rather than episodic consulting engagements. Companies like Bristol-Myers Squibb and CVS Health have concluded that their internal teams, with deep domain knowledge and continuous engagement, are better positioned to identify and scale valuable AI use cases than external advisors.
• Organizations should resist the fear of missing out that consulting firms have aggressively marketed. The fact that deployments are proving difficult even for well-resourced companies with consulting support suggests that patience and experimentation may be more valuable than rushed implementations. Leaders should focus on building internal capabilities, identifying high-value use cases, and learning through controlled experiments rather than betting heavily on unproven approaches.

This does not mean consultants offer no value. They can provide useful perspectives on what other industries are attempting, offer additional capacity for specific projects, and help organizations think through strategic implications. However, expecting consultants to lead transformative AI implementations may be unrealistic given the technology's current maturity level and the expertise required.

The Broader Consulting Industry Reckoning

The AI implementation gap exposes deeper questions about consulting's evolution. For decades, major firms have thrived by positioning themselves as trusted advisors who help enterprises navigate complexity and adopt new capabilities. This model has generated enormous revenue and influenced countless strategic decisions.

However, as technology becomes increasingly central to competitive advantage and enterprises build stronger internal technical capabilities, the traditional consulting value proposition faces pressure. Companies are less willing to outsource strategic technical decisions to external advisors who may lack the deep expertise they claim.

The rise of specialized boutique firms, freelance consultant platforms like Catalant, and direct hiring of technical talent all reflect this shift. Enterprises have more options for accessing expertise, and they are increasingly selective about when traditional consulting firms offer genuine value versus expensive overhead.

For consulting firms, adapting to this reality requires difficult choices. Some are acquiring technology companies and building product capabilities to complement advisory services. Others are investing heavily in proprietary AI tools and platforms. Still others are refocusing on their traditional strengths in strategy, operations, and change management while de-emphasizing technology implementation.

The firms that succeed will likely be those that honestly assess where they can deliver unique value versus where clients are better served by other approaches. Promising to lead AI transformations they cannot yet deliver damages credibility and accelerates clients' decisions to build internal capabilities.

Looking Forward

The generative AI boom has exposed a fundamental tension in the consulting industry's business model. When transformative technologies emerge, consultants face a dilemma: claim expertise to capture the opportunity, risking credibility when clients discover the gap between promises and reality, or honestly acknowledge limitations, potentially ceding ground to competitors and losing market relevance.

Most firms have chosen the former approach, wagering billions on aggressive marketing and capacity building. Clients are now rendering their verdict, and it is largely negative. The companies ending engagements and bringing AI work in-house are not aberrations but potential harbingers of a broader shift.

This does not spell doom for the consulting industry. These firms have weathered technology disruptions before and possess substantial resources, relationships, and capabilities. However, it does suggest that their role in the AI revolution may be more limited than they projected, with implications for both their business models and client strategies.

For business leaders, the lesson is clear: external advisors can provide value, but transformative technology deployment increasingly requires strong internal capabilities, sustained organizational learning, and realistic expectations about expertise availability. The consulting industry's struggles with AI implementation are a reminder that in periods of genuine technological disruption, everyone is learning together, and claims to superior knowledge should be carefully scrutinized.

The AI boom is not leaving consultants behind because they lack intelligence or effort. It is leaving them behind because the traditional consulting model fits poorly with technologies that are genuinely novel, rapidly evolving, and require deep integration with organizational capabilities. Recognizing this mismatch is the first step toward more realistic expectations and better decisions about where external advisory support can genuinely add value.

As generative AI matures over the coming years, consulting firms may indeed find their opportunity in the second wave that industry observers predict. Until then, both consultants and their clients must navigate an uncomfortable period where the gap between promise and delivery remains substantial, forcing difficult conversations about value, expertise, and the future of advisory relationships in an age of technological transformation.