Why the AI Adoption Gap Not Technology Itself Will Define Our Future

Zack Kass and the question leaders should be asking

Zack Kass wants business leaders to stop asking the wrong question about artificial intelligence. As the former head of Go-To-Market at OpenAI, Kass has spent years watching executives fixate on what machines can do while ignoring the more important issue: what we will allow them to do. In his new book, The Next Renaissance: AI and the Expansion of Human Potential, Kass presents a provocatively optimistic thesis that challenges the prevailing narrative of AI anxiety. Yet his argument—while compelling in its historical parallels and reframing of the work crisis—reveals as much about the challenges ahead as it does about the opportunities.

The “adoption gap”: a better frame for AI strategy

The central concept Kass introduces is the adoption gap—the space between the technological threshold (what technology can do) and the societal threshold (what we want it to do or permit it to do). This framing represents a crucial shift in how business leaders should think about AI strategy. Rather than racing to implement every possible automation, organizations must navigate the far more complex terrain of social acceptance, regulatory constraints, and cultural readiness.

The Renaissance metaphor: useful, but incomplete

The Renaissance metaphor that anchors Kass’s argument provides useful historical context but also warrants scrutiny. He draws parallels between our current moment of AI anxiety and the late Middle Ages, a period when most people had good reason to believe the world was deteriorating. Disease, declining populations, and regression in freedom and safety characterized the era immediately preceding explosive growth in art, science, mathematics, and politics. The implication is clear: our current fears about AI may similarly precede an era of unprecedented human flourishing.

Yet this metaphor glosses over inconvenient historical realities. The Renaissance unfolded over centuries, not years or decades. It benefited a narrow elite while millions remained in poverty. The printing press, often cited as the period’s transformative technology, triggered religious wars and social upheaval alongside its democratization of knowledge. The parallel suggests a bumpy road ahead, even if the destination proves beneficial.

What research suggests about jobs, disruption, and transition pain

Research from MIT’s Work of the Future initiative provides empirical grounding for both optimism and caution. Their comprehensive analysis shows that while automation has historically created more jobs than it destroyed, the transition periods have proven brutal for displaced workers. The speed of AI adoption may compress these transition periods dramatically, creating adjustment challenges that dwarf previous technological disruptions.

From job displacement to identity displacement

Kass’s most provocative contribution may be his reframing of job displacement as identity displacement. This psychological insight cuts through economic abstractions to address the real source of AI anxiety. People don’t fear unemployment primarily because of lost income but because work provides meaning, structure, and social connection. The question is not whether there will be work to do but whether that work will provide the same identity anchors that previous generations took for granted.

This reframing has profound implications for organizational change management. Leaders implementing AI cannot treat it purely as a technical challenge or even an operational one. The transformation requires attending to the emotional and psychological dimensions of work redesign. Companies that acknowledge the identity crisis while providing pathways to new forms of meaningful contribution will navigate the transition far more successfully than those that treat workers as interchangeable units of labor.

The apprenticeship problem and the coming knowledge-transfer gap

The apprenticeship problem Kass identifies deserves particular attention from business leaders. In fields from law to consulting to finance, junior employees traditionally spent years doing grunt work that, while tedious, built deep understanding of how organizations function. As AI automates these tasks, organizations face a dangerous knowledge-transfer gap. Future leaders may arrive in senior positions without ever having done the foundational work that built institutional wisdom in previous generations.

Consider the implications for professional services firms. When AI handles document review, financial modeling, or preliminary research, how do associates develop judgment about what matters? The hotel management analogy Kass uses resonates: effective leaders typically worked their way up through every role, building appreciation for each function. Strip away that developmental pathway, and you risk creating executives who understand efficiency but not effectiveness—who can optimize existing processes but cannot reimagine them.

Some organizations are already grappling with this challenge. Leading law firms are redesigning associate development programs to focus on higher-order skills from day one. Consulting firms are creating new apprenticeship models that pair junior staff with AI tools on complex analytical challenges rather than routine tasks. The firms that solve this puzzle will build sustainable competitive advantage through superior talent development.

Why society judges machine failures differently than human errors

Kass’s discussion of the adoption gap’s drivers reveals the complexity of technology diffusion. We tolerate human error far more readily than machine failure, creating asymmetric standards that slow AI adoption even when machines demonstrably outperform humans. This dynamic plays out across industries. Radiologists who miss diagnoses face professional consequences but rarely public outrage. An AI system with a better overall accuracy rate but occasional spectacular failures becomes front-page news and regulatory fodder.

Productivity, inequality, and the distribution problem

Research by Daron Acemoglu at MIT challenges some of Kass’s more optimistic assumptions about AI-driven productivity growth. Acemoglu’s work shows that thus far, AI has failed to deliver the transformative productivity gains that technologies like electricity or computing provided in previous eras. Moreover, AI may be increasing inequality by disproportionately benefiting capital over labor and skilled workers over less-skilled ones. The distribution problem Kass mentions may prove far more intractable than his framing suggests.

Three AI automation scenarios leaders should plan for

The three scenarios Kass outlines for how AI automation might unfold deserve careful consideration by business strategists:

• Massive automation leading to an economic boom followed by distribution challenges. Winners will reap enormous gains while others struggle, creating political pressure for redistribution. Companies should prepare for heightened scrutiny about the social impact of their automation decisions.
• Minimal automation and economic stagnation, which might occur in many sectors due to regulatory constraints, social resistance, or simple risk aversion. Organizations in healthcare, education, and government may find the societal threshold particularly resistant to movement, creating persistent inefficiencies and competitive disadvantages compared to less regulated sectors.
• Asymmetric automation with protected enclaves, which may be the most likely and most problematic. Some industries will race ahead with AI adoption while others remain frozen by regulation or cultural resistance. This creates market distortions, inequality between sectors, and political tensions.

Business leaders must navigate not just their own industry’s adoption trajectory but also the cross-sector implications of uneven technological diffusion.

Near-term risks: alignment, explainability, and bad actors

On the question of risks, Kass makes an important contribution by moving beyond abstract existential threats to concrete near-term dangers. The alignment problem he describes is not about ensuring AI values humanity in some philosophical sense but about the impossibility of universal value alignment in a diverse world. An AI aligned with one culture’s values is necessarily misaligned with another’s. This reality has significant implications for global companies deploying AI systems across different cultural contexts.

The explainability challenge Kass raises touches on a fundamental tension in AI development. The most powerful AI systems are often the least interpretable. As organizations deploy these black-box systems in consequential domains, they face growing demands for transparency from regulators, customers, and employees. Yet requiring explainability may mean sacrificing performance. This tradeoff will define many strategic decisions about AI deployment in the coming years.

Kass’s warning about low- and medium-resource bad actors represents perhaps the most concrete near-term risk. As AI capabilities democratize, individuals and small groups gain unprecedented power to cause harm through fraud, theft, disinformation, or even bioweapon development. Companies must invest far more seriously in security and resilience against AI-enabled threats than current spending patterns suggest.

Advice for workers—and the responsibility of employers

The advice Kass offers to younger workers—focus on learning how to learn and develop resilience through mastering challenging skills—has merit but may be incomplete. Research from Oxford’s Carl Benedikt Frey suggests that workers most at risk from automation are precisely those with fewer resources to invest in continuous learning. The learning-focused solution works for knowledge workers with educational privilege but offers little comfort to those in routine cognitive or manual roles.

Organizations bear responsibility for creating learning pathways for all workers, not just elite talent. Companies that invest in broad-based reskilling programs will build more resilient workforces and stronger social licenses to operate. Those that adopt a Social Darwinist approach—letting workers sink or swim in the new environment—will face growing backlash and regulatory constraints.

The deepest challenge: meaning, purpose, and human dignity

The cultural and spiritual decay Kass mentions—the risk that humans outsource all critical thinking and subsist comfortably but meaninglessly—may be the deepest challenge AI presents. If machines can do everything we once valued ourselves for doing, what becomes of human purpose and dignity? This is not a question technology can answer. It requires philosophical, cultural, and potentially spiritual responses that business leaders often feel uncomfortable addressing.

Yet organizations cannot avoid this existential dimension of AI adoption. When work loses meaning, engagement collapses, creativity withers, and humans become passive consumers rather than active creators. Companies that help employees find new sources of meaning and contribution in an AI-augmented world will vastly outperform those that treat workers as costs to be minimized.

Speed, policy, and why history may not repeat

Kass’s optimism about human adaptability rests on historical precedent—we have repeatedly found new forms of valuable work as old work disappeared. But the speed and comprehensiveness of AI-driven change may break historical patterns. When horses were replaced by automobiles, horses did not retrain for new careers. If AI genuinely replicates broad cognitive capabilities, the analogy to previous automation waves may be less relevant than optimists assume.

The timeline question also demands attention. Kass suggests that adoption gaps will close faster than historical precedent because we lack policy constraints of previous eras. Yet regulatory frameworks are emerging rapidly around AI, potentially slowing deployment significantly. The European Union’s AI Act, China’s algorithmic governance rules, and emerging U.S. frameworks suggest that policy will indeed constrain technological possibility, just as it did during the Renaissance.

Strategic imperatives for business leaders

For business leaders navigating these uncertainties, several strategic imperatives emerge from Kass’s analysis:

• Invest in understanding societal thresholds as much as in advancing technological capabilities. Brilliant AI implementations fail when they clash with cultural values or regulatory requirements. Organizations need sociologists and ethicists on AI teams alongside engineers and data scientists.
• Address identity and meaning explicitly in work redesign. Change management programs that focus only on skills training while ignoring psychological impact will produce resistance and disengagement. Help employees envision not just what they will do differently but who they will become in the transformation.
• Redesign apprenticeship and development pathways for a world where AI handles routine work. How will the next generation of leaders develop judgment and institutional knowledge?
• Prepare for asymmetric automation across sectors and geographies. Strategic plans must account for competitors, suppliers, and customers moving at different speeds.
• Take near-term risks seriously—alignment, explainability, and bad actors—by building security, transparency, and ethical frameworks into systems from the start.

Conclusion: the adoption gap is where outcomes are decided

Kass’s Renaissance metaphor ultimately proves both inspiring and cautionary. Yes, transformative technologies can unlock human potential and drive progress across multiple domains. But that progress is neither automatic nor evenly distributed. The Renaissance benefited Europe while much of the world experienced colonization. The Industrial Revolution created wealth while imposing brutal costs on workers who became, in Karl Marx’s phrase, appendages to machines.

The question is not whether AI will transform work and society but whether that transformation will prove broadly beneficial or narrowly extractive. The adoption gap Kass identifies is precisely the space where that determination will be made. Organizations that approach it thoughtfully—attending to human dimensions alongside technical possibilities—will help shape a future that expands rather than contracts human potential.

The stakes could not be higher. As Kass notes, we stand at an inflection point where choices made in the next decade will shape trajectories for generations. Business leaders cannot afford to be either naively optimistic or reflexively pessimistic. The challenge requires clear-eyed assessment of both risks and opportunities, combined with genuine commitment to creating futures where technology serves human flourishing rather than replacing it.

Kass’s call for humility deserves particular emphasis. We are remarkably poor at imagining unknown possibilities, and the future will likely include developments we cannot currently fathom. That uncertainty cuts both ways—it should temper both doom predictions and utopian promises. What remains certain is that human choices, not technological determinism, will define whether AI enables a renaissance or something far darker. The adoption gap is where those choices get made, one decision at a time, by leaders in organizations around the world.