The rapid integration of artificial intelligence into the modern workplace has created a striking divergence in professional outcomes, according to new research that identifies a growing "capability paradox." While senior executives and high-level decision-makers report significant personal gains in productivity, creativity, and work quality, their junior counterparts are finding it increasingly difficult to derive the same value from these transformative tools. This gap suggests that rather than acting as a universal equalizer, AI may currently be widening the divide between experienced veterans and those just entering the workforce.
According to a comprehensive study recently released by PA Consulting, the benefits of generative AI are being felt most acutely at the top of the corporate ladder. Nearly three-quarters—approximately 70%—of senior leaders surveyed revealed that AI has empowered them to achieve goals and complete tasks that were previously beyond their reach. For these seasoned professionals, the technology is not merely a tool for automation but a sophisticated partner that augments their existing expertise, allowing them to push the boundaries of their roles.
The research paints a picture of a leadership class that has successfully integrated AI into the rhythm of their daily professional lives. On average, senior figures are engaging with AI tools 6.9 times per day. This frequent usage is translating into tangible improvements in output, with 72% of leaders reporting that they are delivering higher-quality work than they did before the AI revolution. Furthermore, more than two-thirds of these respondents (68%) noted that the technology has effectively liberated them from the "drudgery" of repetitive, administrative tasks, freeing up critical mental bandwidth to focus on high-value strategic initiatives.
However, beneath this veneer of executive success lies a more complex and troubling reality for the broader enterprise. The PA Consulting report warns of a "capability paradox," wherein the very individuals who might benefit most from a productivity boost—entry-level and junior employees—are the ones seeing the fewest gains.
Entry-level workers need a helping hand
The disparity between the C-suite and the cubicle is stark. While senior leaders are thriving, only 38% of entry-level or junior employees reported similar benefits in terms of efficiency or task completion. The quality gap is equally pronounced; only 43% of junior staff felt that AI was helping them produce better work, a significant drop compared to the 72% seen among leadership.
Dean Collins, a people and technology expert at PA Consulting, suggests that this paradox is rooted in the relationship between human experience and machine output. Speaking to ITPro, Collins explained that AI essentially "presents back to the individual their own capability and expertise." In this framework, the more experience a professional brings to the table, the more effectively they can leverage AI to enhance their work.
"The more senior someone is, the more likely they are to use AI, and the more confident they are likely to be in their own skills," Collins told ITPro. He noted that a professional with decades of experience already knows "what good looks like." They possess the institutional knowledge and the industry-specific intuition required to interrogate AI-generated content, challenge its assumptions, and refine its outputs until they reach a robust, reliable conclusion.
For a senior leader, AI acts as a sophisticated sounding board. They can use it to test hypotheses or synthesize vast amounts of data, confident in their ability to spot a hallucination or a logical flaw. For a junior employee, however, the technology can be more of a mystery. Without a deep well of experience to draw from, entry-level workers may struggle to provide the nuanced prompts required for high-quality results, or worse, they may lack the critical eye necessary to vet the AI’s work.
This lack of confidence is reflected in the study’s findings on trust and accountability. Only one-in-eight respondents (12%) said they trust an AI’s judgment over their own. Perhaps more tellingly, more than a quarter of workers (26%) admitted they feel uncomfortable signing off on work that has been shaped by AI. This discomfort often stems from a fear that they cannot fully verify the accuracy or the "soul" of the work being produced.
To bridge this widening gap, Collins argues that organizations must move beyond simply providing access to AI tools and start focusing on development. The study outlines a strategic four-point approach to narrowing the capability gap, with a heavy emphasis on skills development for those in the early stages of their careers.
"The main one that I mention is development for people in more junior roles, because we, or many organizations, are giving AI to everyone," Collins noted. He warned that if junior staff are not equipped with the specific know-how to interrogate AI outputs, they will remain unable to unlock the technology’s full potential.
The implications of this gap extend far beyond current productivity metrics; they threaten the very future of the talent pipeline. Historically, junior roles have served as a "proving ground" where employees learn through trial, error, and repetitive tasks. As AI takes over these foundational duties, there is a risk that the leaders of tomorrow will miss out on the formative experiences that build professional intuition.
"If we look five or 10 years into the future… in a decade or so, they’re going to be leaders of the future," Collins told ITPro. "As they progress through their career, if AI is used more extensively, then is AI potentially preventing them from failing and preventing them from learning, which are then the things you lean on when you’re a leader?"
Human expertise in the spotlight
The "capability paradox" highlighted by PA Consulting serves as a powerful reminder that while the technology is groundbreaking, the human element remains the decisive factor in its success. AI skills development has become a central theme for enterprises over the last three years, with business leaders worldwide scrambling to bolster workforce capabilities. However, the nature of these "skills" is shifting.
The PA Consulting report suggests that the rise of AI is actually making traditional human expertise more valuable than ever. The ability to provide context, exercise ethical judgment, and apply critical thinking is what separates a generic AI output from a high-quality professional deliverable. "The technology that promises to put expertise within everyone’s reach may ultimately make human expertise more valuable than ever," the report concluded.
This sentiment is echoed by other major industry studies. Research from Multiverse last year found that "soft skills" have become a primary focus for business leaders looking to integrate AI. Rather than focusing solely on technical proficiency or coding, enterprises are increasingly adopting a "blended" approach that combines AI literacy with broader development programs in communication, leadership, and problem-solving.
Furthermore, recent analysis from Kyndryl’s third annual Readiness Report underlined the urgency of this mission. Kyndryl warned that many firms are currently failing to invest sufficiently in skills development, a failure that is actively harming AI adoption rates. Without a workforce that feels confident and competent using these tools, the massive investments companies are making in AI infrastructure may fail to yield a return.
The challenge for modern organizations is to create a culture where AI is used to enhance learning rather than replace it. This involves encouraging experimentation among junior staff and providing them with the "safety net" needed to fail and learn in an AI-augmented environment. By boosting AI competency early on, firms can ensure that their future leaders are not just passive users of technology, but master practitioners who can direct it with confidence.
Ultimately, the PA Consulting study suggests that the "capability paradox" is a hurdle, not a permanent barrier. By recognizing that AI’s effectiveness is tied to human experience, businesses can pivot their training strategies to focus on the junior and mid-level managers who will eventually steer the ship. In doing so, they can ensure that the productivity gains currently enjoyed by those at the top are eventually shared across the entire enterprise, securing both current efficiency and future leadership.
