As leaders in technology, media and finance gather in Sun Valley this week, much of the conversation will center on AI – its incredible pace, its commercial potential and its implications for nearly every industry. But beneath the excitement lurks a quieter strategic question that may ultimately be more important: If AI democratizes knowledge, what becomes the source of sustainable competitive advantage?
The answer is not simply better technology. It is the best expertise. AI excels at analytical thinking and complex data integration. It can synthesize large amounts of information, identify patterns across complex data sets, augment human judgment, and even simulate empathy in customer interactions. Skills that only recently differentiated high-performing professionals are rapidly becoming table stakes.
As AI compresses the value of routine knowledge work, the prize shifts to the kind of thinking machines still struggle to replicate: connecting seemingly unrelated ideas, reframing problems, and generating new solutions. This is where deep expertise becomes a strategic asset rather than just a professional credential. Humans still surpass technology in their ability to connect different ideas and see things differently. This is especially true for humans with deep expertise, who consistently outperform non-experts and AI in solving complex problems within their domain.
Their advantage is not simply that they know more. Years of deliberate practice fundamentally reshape the way they think. Experts recognize meaningful patterns faster, retrieve relevant knowledge more efficiently, and evaluate problems through richer and more interconnected mental models. Instead of working out every decision from scratch, they rely on sophisticated internal frameworks that allow them to generate more adaptive and often more innovative solutions.
However, this advantage is very domain specific. As a psychologist Timothy Salthouse observed, an expert is “someone who constantly learns more and more for less and less”. The sophistication of expert thinking can become surprisingly fragile. Performance often declines rapidly as experts move beyond the boundaries of their discipline.
Deep expertise can also create blind spots and limit their thinking. Research shows that for truly novel problems, non-experts sometimes outperform specialists in fields ranging from medicine to forecasting. Roughly 60 percent of disruptive innovations originate outside the industries they ultimately transform, a reminder that new perspectives often challenge assumptions that experts no longer question. This presents one of the defining talent challenges of the AI era. Organizations need experts who continue to learn across borders.
The key to building expertise in an AI world is breadth of experience. Our research shows that exposure to a wide range of challenges helps people build richer mental models by strengthening cognitive skills that allow them to learn from new situations rather than simply relying on past experience. Breadth transforms technical competence into adaptive expertise.
For organizations, this means that talent strategy can no longer focus exclusively on deep specialization. It should deliberately combine depth with breadth. Connecting experts from different disciplines is one way to achieve it. Cross-functional communities of practice allow specialists to borrow knowledge from adjacent fields, exposing them to problems, perspectives, and ways of thinking that they would never encounter within their silos. Companies like Procter & Gamble have long embraced this model of “constructive disruption,” recognizing that innovation often occurs at the intersection of disciplines rather than within them.
Technology can also strengthen these connections. The same AI tools that dominate conversations in Sun Valley may be most valuable in helping experts collaborate more effectively. Companies like HumanCorps are increasingly using new artificial intelligence tools to identify unexpected relationships between experts across functions and accelerate the exchange of ideas. The goal is not to replace expertise with AI, but to use AI to help build expertise.
This challenge becomes even more urgent when considering how expertise is developed in the first place. Much of the repetitive work that historically served as practice for future experts is disappearing as AI automates entry-level tasks. At the same time, too much reliance on AI for research, analysis, and reasoning risks weakening the very cognitive muscles that support deep expertise: critical thinking, pattern recognition, and independent judgment. Without deliberate intervention, organizations may face an expertise gap in a decade, not because talent is unavailable, but because fewer professionals have accumulated the experiences necessary to become true experts.
Organizations can counter this erosion through more intentional talent development. Learning technology companies like TalentXTools are building company-specific business simulations as an engaging way to provide exposure to a host of relevant challenges – a digital substitute for experience. Built-in mechanisms that support triple-loop learning and knowledge gathering accelerate the learning process, making it a more efficient way to build foundational skills and expertise.
Similarly, there has been a renewed commitment by companies to early career talent strategies. For example, SAP is deliberately retooling its talent approach with the specific goal of fostering early career professional expertise to scale innovation. When designing these types of talent programs, we typically look to include job rotations, strategic projects, and immersions to provide breadth, while coaches and mentors serve as learning accelerators.
Data collected through diagnostics injected into both of these processes can also be used to target formal education and skills development with greater precision. This increases the gains in participant engagement, as the formal learning components feel more personalized and relevant.
With expertise as a vital driver of value creation and sustainable competitive advantage, targeted talent programs – both digital and in-person – provide essential scaffolding. They help organizations to enhance sophisticated skills and deep domain knowledge. When combined with mechanisms for connecting the dots across domains and work environments architected for slow thinking and lively collaboration, this deep expertise can complement the speed and efficiency of AI with innovation that enables business leaders to build the future readiness needed to stay ahead.
As executives leave Sun Valley, they will likely have new ideas about AI designs, chips, infrastructure and ways to reduce costs. These conversations are significant, but the best-performing organizations over the next decade may not be the ones with the biggest compute budgets. The long-term differentiator may be something less obvious: whether organizations continue to produce people capable of original thinking. Technology can provide more and more answers. Competitive advantage will accrue to organizations that continue to ask better questions.
Future-ready talent: Building a talent pipeline for sustainable business success BY Tania Lennon and Ric Roy IS released on the 28thth July, published by Kogan Page.
