The New Babel: AI's Impact on Talent and Organizational Future

Organizations are experiencing an unprecedented experiment: for the first time, four generations and an algorithm coexist within structures designed for two, perhaps three dominant management models. From the last baby boomers to Generation Z, through Generation X and millennials[1], each brings not only different workplace expectations but fundamentally distinct cognitive frameworks for processing reality, risk, and reward. To this Tower of Babel of perspectives, we add a sixth actor: artificial intelligence, which is not a neutral tool. It is an amplifier, a mirror, and increasingly evident, a destroyer of the very foundations upon which organizational knowledge was built.

The managerial promise had been to integrate the speed of the young with the wisdom of veterans, multiplied by the analytical capacity of machines. Reality, imposed by the pursuit of productivity, prevails.

An MIT Sloan Management Review study analyzing over 10,000 managers revealed radical differences not in competencies, but in cognitive styles by age. Younger managers prioritize concrete techniques and self-assertion, older ones favor coalition-building and intuition derived from lived experience. These young managers developed their managerial style after years in entry-level positions where they learned fundamentals, made mistakes, and gradually built judgment.

What happens when AI eliminates precisely those entry-level positions? We are now celebrating generational diversity while destroying the very mechanism that makes it possible.

The Architect of Babel

Consider the anonymous builder of the Tower of Babel from religious myth. He was probably the best engineer of his time: he had the most ambitious design ever conceived, abundant resources, and a team motivated by faith. His project didn't fail due to lack of talent or capital. It failed because his workers suddenly stopped understanding each other.

But imagine an additional dimension: what would have happened if, simultaneously with the confusion of languages, the entry-level workers — those who mixed mortar, cut stone — simply disappeared? The masters, speaking incompatible languages, would have found themselves without the next generation to replace them. The tower would not only have stagnated due to incomprehension, it would have collapsed due to the absence of generational renewal.

This is precisely our organizational challenge. Companies not only face a linguistic babel among existing generations but an existential crisis at the base: entry-level positions from which skills and experience were traditionally built are being automated. McKinsey indicates that a significant percentage of work activities requiring natural language understanding can be automated by generative AI, and this is just an estimate at the moment when AI is expanding today.

It is precisely those repetitive but formative tasks where judgment is developed, inexpensive mistakes are made, and intuition is built that decades later manifests as "senior wisdom."

We Learn From What We Experience

Generational differences have specific neurobiological roots. The brain reaches its peak processing speed around age 30 — the cognitive advantage of young managers who process new information quickly and adapt behaviors agilely. But studies by Park and Reuter-Lorenz show that complex pattern recognition and the capacity to synthesize fragmented information continue improving until ages 60 or 70. A different neural architecture, built by experience that reconfigures knowledge organization.

A manager who lived through the 2008 crisis as an established professional manages economic uncertainty differently from someone who experienced the 2020 pandemic as their first workplace crisis. Both possess cognitive resources adjusted by different historical events. The MIT study confirms these differences: younger managers favor concrete techniques and demonstrations of individual competence, older ones prioritize coalition-building and intuitive synthesis developed through decades of observing patterns and experiencing.

AI as a Prism

AI simultaneously democratizes access to knowledge and fragments shared understanding. Research by Brynjolfsson and collaborators at MIT documents how AI enables less experienced workers to achieve productivity levels previously reserved for experts. A study with Boston Consulting Group consultants showed a reduction in the performance gap between novices and veterans by more than 30%.

It's verifiable: AI is an equalizer. But the same data reveal a hidden consequence. When tasks exceeded AI's current analytical capabilities, dependent users performed significantly worse than those with experience. As Melissa Valentine of Stanford observes, we are creating "phantom competencies:" a degree of performance that appears solid due to software but lacks transferable and inheritable foundations.

A junior analyst using AI can produce reports that previously required years of experience. But they're not developing their contextual understanding, situational judgment, and intuition about what questions to ask that were built during those years of "basic" work. When facing problems that AI cannot envision, precisely the most valuable strategic challenges, they will lack the foundations to act.

Research by economists David H. Autor and David Dorn documents how automation eliminated routine jobs but created demand for complementary skills. Generative AI operates at a different level: it doesn't just automate tasks, it automates learning itself. Recent articles published by Mohammad Hossein Jarrahi reveal how workers with extensive AI use develop what he calls "algorithmic dependency:" an inability to operate without AI assistance, not out of convenience but due to atrophy of fundamental skills that were never, and possibly will never be, developed.

Older generations developed their expertise in an era when digitalization primarily focused on creating infrastructure for connectivity and converting analog information, consolidating with the development of technologies that began exploiting data in more advanced ways. Their youthful mistakes were made in tasks that are now automatable.

Younger generations face reproducing a learning curve, now with AI, but at the cost of never building the experiential base that underpins developing the judgment to make decisions from lived experience, not from computer simulations. We are creating "towers with shallow foundations:" structures for executing strategies in organizations that could lack the depth to withstand the very dynamics of accelerated change imposed by technological adoption.

The Contraction of the Generational Contract

The implicit contract between generations was clear: young people accepted learning through routine tasks in exchange for understanding processes and growing; veterans tolerated the inefficiencies of beginners by investing in developing a talent network. This exchange sustained a transfer of knowledge, expertise, and that kind of wisdom that only sweat, fatigue, and the occasional tear of frustration can provide.

AI is compacting this contract without replacing it. Why hire three junior analysts when a senior with AI produces the same result? The logic of productivity and costs is undeniable in the short term, but in the medium term, where will future seniors come from? Will structures be operable with a third or less of management structures?

Generational expectations aggravate this. Deloitte research shows that learning and development is among Gen Z's Top 3 priorities when choosing an employer; they're not attracted to the process of learning through routine tasks. AI solves this by automating the tedious. But this logic ignores that the "tedious" is pedagogical: manually reviewing a hundred contracts teaches nuances that no course can transmit.

At the same time, experienced employees feel threatened by the technological pace. If juniors supported by AI produce good quality work, what justifies their experience and compensation? This anxiety generates defensive behaviors: resistance, complexity creation, or control. The builder of Babel would observe: "When the masters felt that new tools would make them obsolete, they stopped teaching. And when the apprentices relied completely on those tools, they stopped learning. The tower stopped due to the rupture of the system that made its construction possible."

Foundations in Crisis

We don't face a simple problem of intergenerational dialogue. I observe it's about systems, processes, and capabilities. Maintaining the productivity of judgment accompanied by AI, and making performance needs congruent with talent development needs, how much new talent will be required to make the organization of the future sustainable?

First, redefine what each role can transfer to automation through AI agents. If routine tasks will be automated, what impact does this have on positions that manage manual processes and information processing? Some positions may evolve into protected experimentation roles where young people make deliberately pedagogical mistakes, developing judgment that AI cannot transfer. Organizations can provoke exploratory projects where learning is the only KPI. This inverts traditional logic: instead of maximizing efficiency from apprentices, their role is to maximize exposure to complexity where the ability to understand, question, and co-create knowledge and build solutions with intelligent systems is now an essential part of professional preparation.

Second, transform AI into a tacit knowledge trainer. Use it in such a way that the logic of experience becomes processes executed by AI so that seniors make their knowledge visible. Recent research on "collaborative externalization" shows that when experts interact with AI systems — training them, refining their outputs, correcting their errors — they frequently articulate intuitions they had never consciously verbalized. This process of making the implicit explicit benefits both algorithms and observing apprentices. AI-mediated mentorship would amplify the knowledge transfer structure.

Third, create models and processes for "generational reciprocity." Juniors teach seniors about the possibilities of emerging tools and models while seniors help them make decisions from context. This in turn requires a new order within organizations where ignorance is not punished but rapid integration of knowledge is pursued through resolving specific challenges. It's like creating hackathons of challenges where people work in generational peer groups.

These interventions will begin as isolated pilot programs and may evolve into collaboration cells within structures designed to recover and sustain specific knowledge for the needs of technology integration and agility in its management.

Beyond Babel

The builder of Babel failed not because his project was impossible, but because he interpreted language diversity as a problem to solve through imposing a single language rather than a resource to orchestrate. Organizations are close to replicating this error by attempting to homogenize corporate culture or impose AI for effectiveness rather than as an integrating element of generational knowledge. Efficiency and cost is a premise, not a strategy that drives talent decisions.

Continuing with the Babel analogy, AI is fundamentally reconfiguring who participates in construction and what "learning to build" means. When we automate tasks where judgment was traditionally developed, we don't just gain efficiency, we cut elements that connect generations of knowledge.

AI democratizes access to capabilities previously reserved for experts, but simultaneously can prevent new generations from becoming true experts. Like a prism that separates light, AI makes previously opaque knowledge visible but fragments shared understanding into individualized experiences where each person sees their own spectrum without understanding that others see radically different realities.

It is a guaranteed fact that AI will replace entire layers of organizational functions. We only need to make precise decisions not to interrupt development trajectories in such a way that in the near future the need for talent with judgment doesn't provoke loss of competitiveness.

Organizational wisdom resides in the deliberately designed and orchestrated translation space between five generations and, now, the predictive software that appears intelligent.