As board members and senior executives integrate AI strategically and operationally into their businesses, effective leaders tackle five principal barriers to AI adoption to drive ROI and reduce AI risk.
01
Many board members and senior leaders cite organizational skill deficits as the leading barrier to AI adoption in 2026, in terms of both “hard” and “soft” skills. When it comes to hard skills, WTW’s 2025 Artificial Intelligence and Digital Talent Intelligence Report finds that the top five most in-demand digital skill roles globally are software engineer, application developer, data scientist, test engineer and cybersecurity engineer. Newer roles such as full-stack developer, solution architect, machine learning engineer, data engineer, cloud engineer and AI engineer appear in the top 20.
WTW’s 2025 Talent Market Trends Dashboard shows that emerging skills vary by industry and geography. For example, in financial services in Europe, data storytelling, predictive analytics and data visualization are all in the top five emerging skills. In other industries and geographies, prompt engineering, data governance management, cyber threat management, risk modeling and demand generation are listed as top digital skills.
Effective leaders also understand that soft skills are just as important as hard skills. In fact, Cisco CHRO Kelly Jones has blogged that Soft Skills are the New Hard Skills. She observes that skills such as emotional intelligence, critical thinking and problem solving are crucial as AI technology advances. She cited the recent World Economic Forum’s Future of Jobs Report 2025, where the top five core skills for the future are analytical thinking, resilience, empathetic leadership, creative thinking and self-awareness. These skills allow for effective leadership, strategic thinking, agility and change management. Effective leaders build and acquire the skills required for their organizations to implement and derive value from AI.
02
Many board members and senior leaders recognize an underappreciated barrier to successful AI adoption is having the right data, which can be as important as having the right skills. They report that data often are misperceived to be a commodity, while currently the opposite is true. Data availability, ownership rights, regulatory and governance rules and cyber/data security remain challenges, as do accessing, ingesting, cleaning and analyzing data.
Even the most sophisticated AI tools and agents cannot perform as intended without the right data. According to the PEX Report 2025/26, 52% of organizations cite data quality and availability as the primary barriers to AI adoption. Effective leaders build systems to acquire, mine, clean, analyze and safeguard the data required for their organizations to implement and derive value from AI.
03
Board members and senior leaders suggest that it is still early days for AI spending, and many organizations struggle with funding AI investments and addressing trade-offs with other costs. According to a Gartner study, worldwide spending on AI is expected to reach $2 trillion in 2026, with the cost of AI services alone reaching $325 billion.
Without thoughtful capital allocation, AI spending strains other corporate investments, including traditional R&D, M&A, marketing, and hiring and staffing. For example, reports in early 2025 indicated companies slowed hiring due to replacing jobs with AI. However, it became clear later in the year that hiring slowdowns had more to do with businesses funding significant investments in AI through across-the-board (or aggregated) labor cost savings rather than with specific job replacement. Effective leaders take the long-term view when funding AI projects, balancing the ability to achieve short-term ROI and long-term business profitability and growth.
04
Board members and senior leaders increasingly find electricity and water as constraints on AI development and application. Goldman Sachs analysts’ base case forecasts that by 2030, overall power consumption from AI data centers will jump 175% from 2023 levels (their previous forecast was 165%).
AI consumes so much energy because training and building large-scale AI models requires significant computational power to run and re-run data through the models many thousands of times, operating 24/7 with increasingly powerful and energy-intensive chips. WTW recently reported that as global demand for data center capacity is fueled by the AI boom, risk exposure has expanded.
Alternative solutions require years to develop and implement. In the meantime, the availability of energy influences decisions over whether to prioritize servers or people, from everyday household and commercial uses to fighting fires and irrigating agricultural crops. Effective leaders use forward-looking models for comprehensive energy and data center risk management, no longer treating the data centers as standalone assets but as part of the critical, interconnected digital infrastructure underpinning their global strategies.
05
Board members and senior leaders have learned that AI adoption will not achieve desired goals in most cases without reimagination of the processes in which it is applied. Many have suggested that while the humans-versus-AI debate can be well-intended, it is misdirected. They report that the most successful implementations reimagine processes with a “best of” approach, considering which processes are best served by AI, humans or, most frequently, both.
Effective leaders know AI tools and agents perform well on specific tasks, but not on broad processes. They also know the agentic web may change aspects of this. However, the most successful implementations start with point solutions that are then woven together to form an effective broader process once the technology, skills and learning have meaningfully advanced.
In general, early implementations suggest generative AI’s strengths including automation of routine processes; pattern recognition at scale, speed and consistency; data analytics and prediction; complex calculations; and simulated reasoning within boundaries. Human strengths include contextual awareness and application, creativity outside of rules, emotional intelligence, ethical and values-based decision making, flexibility in unpredictable or new circumstances and physical skills. Based on a clear understanding of where to focus on automation versus augmentation versus new sources of AI value, effective leaders reimagine processes involving new forms of human and AI partnerships.
Successful AI adoption suggests that leaders who overcome barriers by investing in the right capabilities, treating data as a strategic asset, allocating capital with discipline, securing energy for data centers and reimagining processes end-to-end will be best positioned to unlock sustainable ROI while managing risk.
A version of this article originally appeared on Forbes on February 17, 2026
