The recent announcement by Jack Dorsey, co-founder of Block, that he laid off nearly half of his workforce citing AI’s labor-saving capabilities has sent shockwaves through business circles. This dramatic move, framed as a strategic pivot toward automation, has reignited the long-standing fear that artificial intelligence will inevitably lead to mass layoffs and widespread unemployment. Dorsey’s prediction that “within the next year, I believe the majority of companies will reach the same conclusion and make similar structural changes” has become the latest voice in a chorus of doomsayers who view AI primarily as a job replacement technology. However, this perspective represents a narrow interpretation of AI’s potential impact, overlooking the complex economic and social factors that will determine how this powerful technology reshapes our workplaces and lives. Rather than accepting this deterministic view, we should critically examine whether this path is inevitable or simply one of many possible futures we have the power to shape.
While Jack Dorsey’s decision at Block has made headlines, it’s crucial to recognize that Block, as a highly specialized financial technology firm, is not representative of the tens of millions of diverse businesses operating across the United States economy. The company’s specific business model, competitive pressures, and organizational culture may have made automation seem like the most viable path forward, but this doesn’t translate to universal applicability. Small and medium-sized enterprises, service-based industries, creative sectors, and businesses with strong customer relationships may find that AI enhances rather than replaces human capabilities. Moreover, the choice Dorsey made was not the only option available—many companies are exploring ways to integrate AI that complements their workforce rather than displacing it. This diversity of approaches suggests that the future of AI in the workplace isn’t predetermined but will depend on the specific contexts, values, and strategic priorities of individual organizations.
The concept of “pro-worker AI,” recently coined by a team of MIT economists, offers an alternative vision for how artificial intelligence could transform our workplaces rather than simply eliminating jobs. Unlike traditional automation that seeks to replace human workers, pro-worker AI is designed to enhance the value of existing human expertise and create entirely new tasks and opportunities. This approach recognizes that true technological progress shouldn’t be measured solely by efficiency gains or cost reductions, but by its ability to elevate human potential and create shared prosperity. When AI successfully augments human capabilities, it can lead to increased productivity, higher wages, and the expansion of employment into new, previously unimaginable fields. This represents a fundamental shift in how we think about technology’s role in the economy—from a tool that displaces labor to one that complements and amplifies human ingenuity.
Concrete examples of pro-worker AI in action demonstrate how this approach can deliver tangible benefits across different sectors. In China, a food delivery platform implemented a voice chatbot specifically designed to support hearing-impaired delivery workers, enabling them to communicate more effectively with customers and significantly improving their performance and job satisfaction. More speculatively, imagine an AI assistant that helps aircraft maintenance workers acquire the knowledge and skills needed to transition into spaceflight maintenance—a scenario where AI doesn’t replace workers but expands their career horizons. These examples illustrate that while it may be easier to imagine the jobs that AI will destroy, history has consistently shown that technological innovation ultimately creates more opportunities than it eliminates. As evidence, approximately 60% of jobs in 2018 were in occupational specialties that didn’t exist in 1940—a powerful reminder that technological disruption, when managed well, leads to job evolution rather than simple elimination.
A crucial insight emerging from research on AI’s impact is that the technology itself is neither inherently pro-worker nor anti-worker; its effects depend primarily on how leaders choose to deploy it. The same AI tool could theoretically be used to support and empower workers, or alternatively to surveil and sanction them, with dramatically different outcomes for workers’ wellbeing and dignity. This fundamental truth—that intention matters more than technology—challenges the deterministic narratives that suggest AI’s impact is somehow inevitable. Rather than accepting a future where AI inevitably displaces workers, we can consciously design systems that augment human capabilities and create new value. The distinction comes down to whether we view AI primarily as a labor replacement technology or as a collaborative partner that enhances human judgment, creativity, and problem-solving capabilities. This intentional approach requires organizational leaders to make deliberate choices about how AI is implemented, what problems it’s designed to solve, and whose interests it serves.
The rapid pace of AI development presents both opportunities and challenges, suggesting that speed of implementation should be a carefully considered strategic choice rather than an unquestioned imperative. Federal Reserve Governor Chris Waller recently highlighted this issue when discussing AI adoption within the Fed’s operations, including payment processing systems. He appropriately challenged the Silicon Valley ethos of “move fast and break things” as incompatible with institutions that bear significant public responsibilities and must maintain trust. Waller’s observation that “AI systems can amplify errors as quickly as they amplify efficiency” underscores why thoughtful implementation with appropriate guardrails is essential. The Federal Reserve’s approach—building internal AI platforms with coding safeguards and clear boundaries—offers a model for how organizations can benefit from AI’s capabilities while minimizing potential risks. This balanced approach recognizes that while technological advancement shouldn’t be unnecessarily slowed, neither should it carelessly proceed without proper oversight and consideration of consequences.
The Federal Reserve’s cautious approach to AI adoption provides valuable insights for organizations across all sectors. By implementing guardrails and establishing clear boundaries for AI use, the Fed demonstrates how institutions can harness AI’s potential while maintaining accountability and reliability. This approach involves several key practices: developing comprehensive AI governance frameworks that establish clear lines of responsibility; implementing robust testing protocols before deployment; maintaining human oversight of critical decisions; and continuously monitoring AI systems for unintended consequences. The Fed’s experience suggests that organizations should resist the pressure to adopt AI simply because it’s fashionable or competitors are doing so, and instead focus on identifying specific use cases where AI can genuinely enhance value without compromising quality or trust. This thoughtful approach requires investment in both technology and the organizational capabilities needed to manage it responsibly, creating a foundation for sustainable AI adoption that delivers long-term benefits rather than short-term hype.
Government has a crucial role to play in encouraging an AI future that broadly serves people rather than narrowly focused on automation and displacement. Public funding could be strategically directed toward pro-worker AI initiatives and projects that use AI to improve public services, drawing inspiration from successful models like the Defense Advanced Research Projects Agency (DARPA) prize-style funding or the public-private partnerships exemplified by Operation Warp Speed, which rapidly developed COVID-19 vaccines. Such approaches could include competitive grants for companies that develop AI systems that demonstrably enhance worker capabilities and create new employment opportunities; tax incentives for businesses that implement AI in ways that complement rather than replace human workers; and support for research into AI applications that address societal challenges while creating decent work. The goal should not be to slow AI development but to steer it in directions that generate shared prosperity and address pressing social needs, creating a diverse ecosystem of AI applications that serve different purposes and stakeholders.
Several factors may explain why the private market, left to its own devices, might overemphasize AI automation while underinvesting in pro-worker applications. According to MIT researchers, this bias could stem from multiple sources: the first generation of AI tools may naturally focus on automation because these applications are more straightforward to develop and demonstrate immediate cost-cutting benefits; shareholder pressure for short-term financial results incentivizes companies to pursue labor-reducing technologies that boost quarterly earnings; and the historical vision of artificial general intelligence has often emphasized worker replacement as a measure of technological achievement. Furthermore, AI that disempowers workers could strategically reduce bargaining power and shift economic profits away from labor toward capital, potentially explaining why some organizations find automation appealing regardless of its broader economic impacts. Recognizing these biases is essential for developing countermeasures that can encourage a more balanced approach to AI adoption.
The economic implications of AI that empowers versus replaces workers are profound and deserve careful consideration. When AI systems enhance worker productivity rather than simply replacing workers, they can potentially increase wages and expand employment into new areas, creating a virtuous cycle of economic growth and shared prosperity. This approach recognizes that the ultimate goal of technological advancement should be to improve human wellbeing broadly, not just to maximize corporate profits or efficiency metrics. In contrast, an automation-focused approach risks concentrating economic gains among capital owners while displacing workers whose skills become devalued—a scenario that could exacerbate inequality and reduce aggregate demand as displaced workers have less income to spend. The historical evidence suggests that when technology complements rather than substitutes for labor, the overall benefits are more widely distributed and sustainable. This perspective challenges the simplistic narrative that AI will inevitably create a winner-take-all economy, suggesting instead that intentional policy choices and business strategies can lead to more inclusive outcomes.
Despite legitimate concerns about AI’s potential disruptions, it’s important to acknowledge the considerable promise this technology holds for boosting productivity growth and expanding the economic pie. AI has the potential to address complex problems, optimize resource allocation, and unlock human creativity in ways that could substantially increase economic output. A growing economy provides a much better foundation for improving people’s lives than a stagnant or shrinking one, suggesting that we shouldn’t view AI solely through a displacement lens. The key question is how we can harness AI’s potential to broadly benefit society rather than allowing it to primarily serve narrow interests. This requires recognizing that AI is not a monolithic force with predetermined outcomes, but rather a collection of technologies whose impacts will depend on how we choose to develop, deploy, and govern them. By focusing on AI applications that enhance human capabilities and create new value, we can potentially achieve both productivity gains and more broadly shared prosperity.
As we stand at this critical juncture in AI’s development, several actionable steps can help steer toward a future where AI empowers rather than displaces workers. For businesses, this means conducting thorough impact assessments before implementing AI, prioritizing applications that augment human capabilities, involving workers in the design and deployment process, and investing in reskilling programs that help employees adapt to changing requirements. For policymakers, this involves developing regulatory frameworks that ensure AI is deployed responsibly while encouraging innovation, supporting research into pro-worker applications through targeted funding, and strengthening social safety nets to support workers during transitions. For workers themselves, the challenge is to develop adaptability and continuously update skills, focusing on areas where human judgment, creativity, and emotional intelligence provide enduring advantages. The decisions we make today will profoundly shape the AI-enabled economy of tomorrow, and by approaching this transformation thoughtfully and intentionally, we can create a future where technology serves as a catalyst for shared prosperity rather than a source of displacement and inequality.