EU AI Act: How a lack of AI competence for bosses is now becoming a real liability risk

Xpert Pre-Release

Online contact (Konrad Wolfenstein)

Language selection 📢

Published on: June 21, 2026 / Updated on: June 21, 2026 – Author: Konrad Wolfenstein

EU AI Act: How a lack of AI competence for bosses is now becoming a real liability risk

EU AI Act: How a lack of AI expertise for bosses is now becoming a real liability risk – Image: Xpert.Digital

AI is flying blind in management: Why 80% of managers are massively overwhelmed

Those who don't lead today will be led: The approaching end of classical management

Artificial intelligence is permeating everyday business life at a breathtaking pace – but while the technology scales relentlessly, a critical weakness is being revealed in the executive suites.

Recent studies reveal an alarming paradox: Although the vast majority of companies are implementing AI tools, most initiatives fail not because of the technology itself, but due to a lack of leadership skills, inadequate governance, and insufficient change management. Over 80 percent of executives admit to feeling overwhelmed by the rapid pace of AI advancement. Strategic consistency often gives way to operational activism that causes pilot projects to fizzle out. With the sharp regulatory winds of the EU AI Act, this knowledge gap has transformed from an internal obstacle into a tangible, personal liability risk. The following article sheds light on the profound AI skills crisis in management, exposes the dangerous discrepancy between written guidelines and actual practice, and shows how leaders can make the crucial leap from reactive observers to proactive shapers of the AI era.

When leadership becomes a bottleneck: The AI skills crisis in management

Many companies are now intensively engaged with artificial intelligence. However, the path to operational practice is hampered by significant management obstacles: Current studies show that over 80 percent of executives admit that their leadership skills and internal governance are unable to keep pace with the rapid advancements in AI. To close this critical gap between aspiration and actual management practice, targeted skills development at the leadership level is no longer optional – it is a strategic necessity. The difference between reacting and acting today lies in the personal AI competence of managers.

The paradox of the AI era: Adoption without transformation

Artificial intelligence has rapidly permeated everyday business. According to the McKinsey State of AI Report 2025, 88 percent of companies are using AI in at least one business area – an increase of just 20 percent in 2017. Generative AI has nearly tripled in just two years, and according to a survey by Nash Squared and Harvey Nash, 90 percent of technology decision-makers worldwide now say they are either piloting or implementing AI at scale – up from 59 percent the previous year.

But behind these impressive figures lies a profound paradox: adoption is not the same as transformation. Only 38 percent of companies have actually scaled AI beyond initial pilot projects. And the gap between technological deployment and organizational maturity is widening instead of narrowing. Two-thirds of companies report that the return on investment from AI pilot projects is currently impossible to measure. The key finding is: the technology is there. Management often isn't.

The figures from Germany are even more striking. In a joint study by the Stifterverband and McKinsey, which surveyed more than 1,000 executives with personnel responsibilities, 86 percent stated that their company could utilize the potential of AI much more effectively. Seventy-nine percent cited the lack of necessary skills among the workforce—and, implicitly, within management itself—as the main obstacle. Instead of strategic consistency, operational activism dominates: AI tools are introduced, but not truly integrated into processes, decision-making paths, and corporate culture.

The governance gap: guidelines on paper, chaos in practice

Perhaps the most troubling finding of recent research concerns not the technology itself, but the governance surrounding it. According to a benchmark study by the AAA-ICDR Institute, which surveyed 500 senior legal and executive leaders from companies with more than $100 million in annual revenue, 87 percent of all companies have now established formal AI governance principles or policies. However, only 22 percent of these companies report that these structures actually work in practice. 56 percent describe their governance as structurally sound, but inconsistent in execution. Another 20 percent see a significant gap between written policies and day-to-day practice.

This picture is confirmed by the 2025 AI Governance Benchmark Report: 80 percent of organizations are already using AI operationally, but only 14 percent have an enterprise-wide AI governance framework. According to a Deloitte study, nearly two-thirds of all organizations implemented generative AI without first establishing adequate governance controls. In a separate 2024 survey by the IAPP, only 28 percent of companies had formally defined oversight roles for AI governance.

The OneTrust study, which surveyed 1,250 IT decision-makers in Europe and North America, reinforces this finding: 82 percent of respondents confirm that AI risks are accelerating the need to modernize governance structures. Nevertheless, virtually all companies are significantly lagging behind. Governance exists as a concept, but its implementation remains the exception.

What does this mean in business reality? Companies that use AI without clear lines of responsibility risk accumulating liability, facilitating data breaches, and losing the trust of customers and regulators. The responsibility for this clearly lies with management – and most currently only bear it on paper.

The blind spot in everyday leadership: Personal AI competence

Anyone tasked with leading companies into the AI era needs more than strategic acumen. They need a sufficiently deep understanding of how AI systems function, where they are reliable and where they are not, what organizational implications their use has, and what legal boundaries must be observed. But this is precisely where a serious deficit exists.

According to the Cambridge Judge Business School Executive Education Survey from October 2024, which polled 200 senior executives, only 12 percent of respondents described their own senior leaders as very well prepared to deal with AI and digital transformation. Thirty-one percent rated their leadership as either rather or completely unprepared. In a survey by Egon Zehnder, only 20 percent of executives surveyed stated that their company possessed the necessary skills to cope with the changes brought about by AI over the next five years. These figures are not self-criticism—they are warning signs.

The picture painted by a more recent data point is hardly better: According to a survey by Akkodis of 2,000 executives worldwide, managers' confidence in their own AI strategy fell from 69 percent in 2024 to 58 percent in 2025. CEOs experienced the steepest decline of all hierarchical levels, at 33 percent, followed by CTOs with a drop of 20 percent. Only 55 percent of executives believe their teams fully understand the risks and opportunities of AI. It's a rare phenomenon: The more experience is gained with AI, the greater the awareness of one's own shortcomings becomes.

The Hernstein Management Report, a representative study of 1,600 executives in Austria and Germany, further reveals that while around 90 percent of executives consider building AI expertise necessary to properly assess the limitations of AI, only 8 percent of companies had internal AI guidelines in place at the time of the survey. The gap between understanding and action could hardly be greater.

Why management is structurally overwhelmed

The AI skills crisis in management is not an individual failure. It is the result of a structural imbalance between technological dynamism and organizational inertia. AI evolves in months; building the necessary skills within companies takes years. Managers were socialized in a world where a basic understanding of technology was not a core requirement for management roles. The decision of whether AI is appropriate in a given context could be delegated to IT departments or external consultants. This delegation no longer works in the age of AI.

A study by IESE Business School and KU Leuven, based on the analysis of 375 million US job postings between 2010 and 2022, arrives at a clear conclusion: AI will not replace managers, but it will demand a fundamentally different style of leadership. Companies with AI systems are increasingly seeking managers, and at the same time, the required skill set is changing radically. Cognitive and interpersonal skills such as collaboration, creativity, stakeholder management, and data analysis are gaining importance, while the need for routine administrative tasks is decreasing. Managers are still needed—but different managers than those currently found in many companies.

The Robert Half study, which surveyed 2,025 C-level executives from 13 countries, paints a picture of an approaching turning point: 84 percent of the executives surveyed see AI as the most important skill requirement by 2035. Around half assume that leadership will be AI-supported and data-driven in the future – but will still need to be guided by human values and intuition. This insight is correct, but it alone does not close the skills gap.

To make matters worse, while many companies offer AI training, they often fail to tailor it effectively to specific target groups. A 2025 General Assembly study of 651 business leaders in the US and UK reveals that while 62 percent of executives reported having participated in AI training (compared to 42 percent in 2024), less than half (47 percent) reported that their company offers AI training specifically for leadership. General AI training for the workforce is often considered sufficient, but what leaders need is a different skill set: not the operation of tools, but the strategic evaluation, integration into governance structures, and management of organizational transformation processes impacted by AI.

Failure starts at the top: Why AI projects so often fail to deliver

Recent studies paint a sobering picture of the success rate of corporate AI projects. According to analyses by the RAND Corporation and McKinsey, 80 percent of AI initiatives fail to deliver measurable business value – twice as many as traditional IT projects. Gartner and RheoData report that 70 to 85 percent of all AI projects fail or never reach full implementation. Only about 30 percent of all AI initiatives even make it past the pilot phase.

The key finding: Failure rarely lies in the technology itself. It lies in how companies approach—or fail to approach—change management. A McKinsey analysis of the most successful AI companies shows that the 26 percent of companies that actually generate tangible added value take a counterintuitive approach: They invest 70 percent of their resources in people and processes, pursue only half as many opportunities as their competitors, and view AI transformation as an organizational capability—not a technology project. The result: 1.5 times higher revenue growth and 1.6 times higher shareholder returns than the competition.

The McKinsey analysis identifies a clear pattern among AI high performers: Leaders actively drive AI adoption, are three times more likely to be directly involved than in other companies, and communicate a clear vision. More than half of these companies have fundamentally adapted their processes to AI, compared to only 20 percent on average. Another indicator of the link between leadership engagement and AI success: Leaders who actively encourage their teams to use AI are seven times more likely to work in companies that offer regular AI training. Leadership personality and learning culture are not independent variables.

Studies show similar patterns at the level of transformation capability in general. Kienbaum's study on the transformation capability of German companies in 2025 concludes that around 70 percent of all transformation projects fail or do not achieve their goals. Key reasons cited include: inadequately prepared managers, deeply entrenched outdated structures, and a lack of consideration for corporate culture. Introducing AI without preparing management for the new level of responsibility doesn't just increase this failure rate—it doubles it.

A new dimension of digital transformation with 'Managed AI' (Artificial Intelligence) - Platform & B2B solution | Xpert Consulting

A new dimension of digital transformation with 'Managed AI' (Artificial Intelligence) – Platform & B2B solution | Xpert Consulting - Image: Xpert.Digital

Here you will learn how your company can implement customized AI solutions quickly, securely and without high entry barriers.

A managed AI platform is your all-inclusive, worry-free solution for artificial intelligence. Instead of dealing with complex technology, expensive infrastructure, and lengthy development processes, you receive a ready-made solution tailored to your needs from a specialized partner – often within just a few days.

The key advantages at a glance:

⚡ Rapid implementation: From idea to ready-to-use application in days, not months. We deliver practical solutions that create immediate added value.

🔒 Maximum data security: Your sensitive data stays with you. We guarantee secure and compliant processing without sharing data with third parties.

💸 No financial risk: You only pay for results. High upfront investments in hardware, software, or personnel are completely eliminated.

🎯 Focus on your core business: Concentrate on what you do best. We take care of the entire technical implementation, operation, and maintenance of your AI solution.

📈 Future-proof & scalable: Your AI grows with you. We ensure continuous optimization and scalability, and flexibly adapt the models to new requirements.

More information here:

The Managed AI Solution - Industrial AI Services: The Key to Competitiveness in the Services, Industry and Mechanical Engineering Sectors

Closing the AI leadership gap: How to keep your company competitive

Regulatory pressure: The EU AI Act as a catalyst

The EU AI Act will be fully applicable from August 2, 2026, having already entered into force on August 1, 2024. This regulation represents the world's first comprehensive legal framework for regulating artificial intelligence and, in principle, affects every company that develops, sells, or uses AI systems. Potential penalties for violations can reach up to €35 million or seven percent of annual turnover. According to Bitkom, one in three companies in Germany already uses AI productively, but very few have fully clarified their compliance obligations.

What is often overlooked is that the EU AI Act explicitly states that companies and other institutions operating AI systems must ensure a sufficient level of AI competence. This is not a voluntary recommendation – it is a legal obligation. Since February 1, 2025, companies have been required to demonstrably train their employees on the AI systems they use. For managers, this means that risk management, documentation requirements, transparency, and compliance with data protection and ethical guidelines are no longer abstract governance issues, but rather personal responsibilities with potential liability implications.

The European Commission estimates compliance costs for individual high-risk AI systems at between €6,500 and €400,000. High-risk systems—such as AI applications in personnel decisions, lending, or medical diagnosis—affect an estimated five to fifteen percent of all AI applications. Those who haven't established AI governance today will pay the price tomorrow not only with lost efficiency but also with concrete fines.

The Deloitte Governance of AI Survey 2025, which surveyed 695 board members worldwide, including 49 from Germany, reveals that AI is still not a topic of discussion on nearly a third of all boards globally – though in Germany, this figure is only 16 percent, representing a significant improvement over the global average. Two-thirds of respondents attest to their board's lack of knowledge and experience in this area. The regulatory window is closing – and most supervisory boards are still outside it.

The skills gap as a systemic risk for Germany

For the German economy, the management skills crisis in the field of AI has a particular strategic dimension. Germany is competing for technological leadership in a globalized economy. Large companies from the US and Asia have made enormous investments in AI infrastructure and expertise. The productivity potential of AI is immense: McKinsey estimates that companies can increase their productivity by almost 20 percent by automating repetitive tasks, streamlining processes, and fostering data-driven innovation.

The TÜV continuing education study 2026, for which 500 decision-makers were surveyed, shows that while three out of four companies (75 percent) offer continuing education opportunities to their entire workforce, only 29 percent have a formally documented continuing education strategy. 65 percent of employers allocate a maximum of €1,000 per employee per year for training. Given the complexity and dynamic nature of AI, this is a structurally inadequate level of investment.

The Slalom study from 2025 adds: In Germany, 55 percent of companies see skills gaps as the biggest obstacle to company-wide AI implementation. A further 47 percent cite employee mistrust and job insecurity as the second biggest barrier. Both factors share a common root: the lack of a clear, competent leadership level that credibly embodies and communicates the AI transformation.

An analysis by the IBM Institute for Business Value further reveals that nearly half of the surveyed executives admit their employees lack the necessary AI knowledge and skills to deploy AI technologies at scale. The World Economic Forum estimates that 50 percent of the global population urgently needs new skills to meet the evolving business demand for AI. This figure could rise to 90 percent by 2030 if nothing changes.

The situation is particularly serious for German SMEs. In one of the most comprehensive surveys of this target group, the AI Study 2025 with 455 AI managers from SMEs and mid-sized companies, 86 percent recognize the relevance of AI as business-critical. However, a lack of expertise, insufficient data quality, and the absence of an AI strategy are cited as key obstacles. While large companies with substantial budgets, internal data teams, and modern infrastructure are taking the lead, SMEs risk falling behind structurally.

From knowledge to action: What constitutes competent AI leadership

The profile of an AI-competent leader can be clearly defined based on the research. It's not about writing code or training models. What's crucial is a strategic and critical understanding that encompasses four levels.

At a conceptual level, leaders need a solid understanding of how different AI technologies work, their strengths and limitations, and the differences between descriptive, predictive, and generative AI approaches. This conceptual foundation is essential for asking relevant questions—to their own team, to suppliers, and to regulators. According to the Cambridge Judge Business School Survey, data-driven decision-making is the most frequently cited key competency for leaders in the AI age, followed by strategic thinking for AI integration.

At the strategic level, the ability to systematically identify, prioritize, and evaluate AI use cases within the company is essential. This includes understanding the ROI logic behind AI investments, recognizing the organizational prerequisites for successful implementation, and critically examining the marketing promises of technology providers. McKinsey demonstrates that companies employing a strategic platform approach and scalable AI deployment achieve ROI values of up to 25 percent – while isolated pilot projects often remain below five percent.

At the governance level, executives are now responsible for defining clear guidelines for internal AI use: Which systems are deployed, who monitors them, how is documentation handled, and how are malfunctions escalated? This is not an IT task. It is a management task. For the enforcement of the EU AI Act, operators—that is, companies that use AI systems developed by others—are explicitly responsible. Compliance cannot be delegated.

Finally, at the cultural level lies one of the most demanding leadership tasks: embedding AI competence throughout the entire organization, building trust in new technologies, and simultaneously addressing legitimate concerns. The Stanford AI Index 2025 reports sales increases of 23 percent, a 31 percent rise in customer satisfaction, and 40 percent faster data-driven decisions where AI has been productively implemented. These results are not achieved through technology alone—they result from a combination of committed leadership, targeted training, and consistent change management.

Xpert.Digital as a partner for overcoming the AI leadership gap

The described skills gap affects managers at all levels – from the executive board and middle management to department heads responsible for AI projects. Closing this gap requires more than individual seminars: it demands a systematic, context-specific, and practical engagement with AI at the leadership level, combined with concrete governance structures that are sustainable in everyday business.

Xpert.Digital supports executives and specialist teams precisely at this intersection. As a digital platform with deep industry expertise in B2B markets, industrial logistics, and digital transformation, Xpert.Digital understands that AI competence is not an abstract educational task, but must always be developed within a concrete economic and regulatory context. The approach combines analytical depth with operational relevance: Which AI applications create genuine added value in which business context? How can governance requirements be implemented pragmatically and in an audit-proof manner? And how can management not only be informed, but also empowered?

The foundation for this is evidence-based content that brings together the latest research, regulatory developments, and practical experience. In a business world where the speed of AI development structurally exceeds the speed of institutional learning, this bridging of research and practice is not a luxury – it is a prerequisite for ensuring that leaders do not become the bottleneck in their own AI transformation.

When waiting becomes a competitive disadvantage

The empirical evidence leaves no room for interpretation: AI is no longer just a topic for tech companies, research institutions, or early adopters. It is a strategic imperative for any company that wants to remain competitive in the coming years. The crucial variable is not the technology itself—it is available and scalable. The crucial variable is management.

A look at the overall data reveals the dramatic nature of the situation: AI adoption stands at 88 percent, but scaled AI use is only at 38 percent. 87 percent of companies have governance frameworks, but only 22 percent function effectively in practice. 90 percent of executives consider AI skills necessary, but only 8 percent of companies have binding internal AI guidelines. 86 percent of German executives believe they are not sufficiently utilizing AI potential, yet over half invest too little in professional development.

This systematic gap between awareness and action is not inevitable. It can be overcome – through a leadership culture that views personal AI competence not as a niche technology, but as a core competency of modern management. The companies that most consistently implement this transformation will not only be regulatory compliant. They will be measurably more productive, innovative, and resilient to market changes.

The difference between reacting and acting, as all data consistently show, lies in the personal AI expertise of executives. Those who invest in this expertise today will gain strategic flexibility tomorrow. Those who wait until external factors force them to act—be it through regulation, competition, or failed projects—pay a significantly higher price. Not in abstract terms, but in euros, market share, and lost years.

🎯🎯🎯 Data-driven B2B industry hub as a quasi-in-house solution

The quasi-in-house solution: How Xpert.Digital closes operational gaps in B2B marketing and sales – Smart Content-Driven Business - Image: Xpert.Digital

Xpert.Digital is a data-driven B2B industry hub led by Konrad Wolfenstein . The company acts as an external, quasi-in-house solution for industrial partners, closing operational gaps in marketing, content, and sales – without requiring additional resources on the client side.

More information here: