The relentless march of artificial intelligence into workplaces across North America is creating a profound identity crisis among employees that no amount of executive reassurance can simply talk away. Research from Harvard Business School reveals that AI adoption poses a triple threat to professional identity—through role compression, control shift, and span erosion—leaving workers feeling diminished just as organisations need their buy-in most. Now, a global study of senior executives shows that more than half are already witnessing de-skilling within their ranks, with over 60 percent warning this will become a material threat within three to five years. The message from both business schools is clear: companies must fundamentally redesign how they introduce AI, placing human identity and capability-building at the centre of their transformation strategies.
The Triple Threat to Professional Identity
Das Narayandas and Shunyuan Zhang, professors at Harvard Business School, have identified three distinct ways AI undermines how employees see themselves professionally.
Role compression occurs when workers find themselves with less meaningful work as AI takes over routine tasks. The duties that remain often feel lower in status because the judgment and expertise that once defined the role have been automated away.
Control shift represents a different kind of erosion. Supervisors who once exercised discretion over decisions that showcased their expertise discover those choices are now delegated to algorithms. The judgment that made the role feel distinctly theirs now sits with the system—even if their job title remains unchanged.
Span erosion completes the picture. As employees increasingly rely on AI for answers and solutions, managers find their influence over people, budgets, and processes gradually shrinking. Their authority becomes more theoretical than real.
Together, these three forces create what the professors describe as a triple threat to professional identity that cannot be dismissed by senior management. Traditional approaches—reassuring employees that everything will be fine—simply will not suffice. Instead, organisations must take deliberate, concrete action to show workers how AI adoption can preserve or enhance their sense of expertise, influence, and professional value.
How Organisations Must Respond
The Harvard researchers outline five practical steps that forward-thinking organisations should implement immediately.
First, rechartering roles. Empty promises that employees will “do more strategic work” ring hollow. Instead, organisations must be specific about which forms of judgment, customer knowledge, interpretation, and relationship expertise will become more prominent as routine tasks are automated.
Second, building decision guardrails. As AI assumes certain decision-making functions, companies must ensure managers and employees retain the power to override and question algorithmic choices. Clearly defined guardrails help counter the identity threat by preserving human agency.
Third, creating analytical overlays. AI should enhance decision-making rather than replace human expertise entirely. “It’s like having a research assistant who can do the background work, but you are still the one bringing meaning to it, so you can feel good about that,” Professor Narayandas explained in Harvard Business School’s Working Knowledge publication.
Fourth, developing redeployment pathways. Companies must commit to retaining and retraining workers who find their roles transformed by AI. Crucially, those new roles must be respected and meaningful—positioned as growth rather than diminishment.
Fifth, ensuring executive sponsorship. When roles are redesigned, senior leadership must publicly champion both the new positions and the people stepping into them. This visible support counters any perception that the changes represent demotion.
“You can’t just turn salespeople into consultants,” Professor Zhang emphasised. “You need the training to provide those pathways, and executive sponsorship to make that happen.”
The De-Skilling Emergency
Beneath the identity crisis lies an equally pressing concern: the very real risk that workers will lose critical skills as AI assumes more of their responsibilities.
A global study conducted by the BCG Henderson Institute surveyed 70 C-suite leaders and senior executives across multiple industries. The findings are striking. Half of those surveyed are already observing de-skilling within their organisations. More worryingly, over 60 percent believe this will pose a material threat to their business within the next three to five years.
The implications extend far beyond individual workers. BCG consultants Sagar Goel, David Martin, and Charikleia Kaffe warn that de-skilling could fundamentally determine an organisation’s future competitive position.
To address this, organisations must actively shape how employees use AI in ways that strengthen rather than supplant human capabilities. This requires defining which AI systems are approved for use, how output verification should be conducted, and crucially, when human review becomes mandatory. Concrete guidance must detail how to use AI in specific contexts while also identifying tasks where AI won’t be used—because originality, ethical judgment, or synthesis remains most critical when performed by humans.
The organisation must fundamentally redesign how work gets done, determining what tasks should be delegated to AI, what tasks should remain human, and where an ensemble approach works best—with humans and AI tackling problems independently before combining their outputs.
“As organisations redesign their workflows, resistance to skill erosion needs to sit alongside efficiency and effectiveness as an explicit design criterion,” the BCG consultants stress.
Rebuilding Human Capabilities
Human skill development must become visible in employee performance reviews, promotion criteria, and development pathways. This means illuminating not just what employees deliver, but how they achieve their outcomes. What skills were exercised? What capabilities were built?
The consultants also urge leaders to establish active routines that help employees rebuild cognitive and creative muscles. “Just as organisations have norms for different meeting types, they need norms that exercise different human capabilities: cognitive warm-ups, AI-free problem-solving sessions, structured debate moments, and reflection prompts.”
One leading bank in India provides a compelling example. The institution runs a structured AI-free session on the first Friday of each month across all business and support functions. These sessions combine case study discussions, design-thinking assignments, and sector briefings—deliberately exercising capabilities that AI cannot replicate.
The underlying philosophy is straightforward: AI transformation is ultimately about humans, and their skills must remain central to any meaningful strategy.
Why it Matters
The way organisations handle AI adoption will define not just their competitive fortunes but the broader social contract between businesses and workers. If companies pursue efficiency gains without addressing the identity crisis and de-skilling threat, they risk creating a workforce that is simultaneously anxious, diminished, and less capable of the creative and strategic thinking that will matter most as AI continues to evolve. The research is unambiguous: reassuring words are not enough. Only deliberate, visible action—from rechartering roles to protecting human agency in decision-making to establishing AI-free capability-building sessions—will determine whether AI becomes a tool that empowers workers or a force that displaces them. In the contest for talent, reputation, and long-term organisational resilience, how businesses treat their people during this transformation may prove more consequential than any efficiency metric.
