Sal Khan, founder of the nonprofit educational platform Khan Academy and a prominent voice on technology and learning, published an opinion piece arguing that artificial intelligence will displace workers on a scale far greater than most people anticipate. Khan’s essay — blending economic analysis, education policy prescriptions and a call for social solidarity — emphasizes that the next wave of A.I. advances will affect not only particularly visible white-collar jobs but large swaths of the workforce across sectors.

From Tutors to Writers: Broad Reach of A.I. Automation

Khan’s argument begins with the nature of recent A.I. systems: increasingly capable models that can perform tasks previously thought to require human judgment, creativity or specialized training. He cites examples including automated writing and customer-service systems, code-generation tools, and models that can draft legal documents, produce marketing materials, or provide diagnostic suggestions in medicine. While earlier waves of automation disproportionately affected manufacturing and routine manual tasks, Khan contends that modern A.I. reaches into cognitive, communicative and decision-making work.

He warns that many professions viewed as secure because they rely on domain expertise or social interaction may still be vulnerable. Khan draws attention to occupations in education, finance, journalism, law, software development and the creative industries, where tools can now produce lesson plans, financial analyses, news summaries, contracts and working code—tasks that previously required substantial human time and training.

Economics of Displacement and the Pace Problem

Khan emphasizes two features of A.I.-driven disruption that make it especially consequential. First is the breadth of tasks that can be automated: rather than replacing a single repetitive action, current models can take on multiple subtasks that together constitute a job. Second is the pace at which these capabilities are improving and being deployed. Khan cautions that adoption curves for software are often steep because of relatively low marginal costs once a system is developed, meaning employers can scale A.I. tools rapidly.

He notes that while some historical transitions gave workers time to retrain or move into different roles, the concentrated and rapid efficiency gains from A.I. could produce significant short-term unemployment and economic dislocation if policy and institutions do not move quickly to buffer workers.

Policy Prescriptions: Education, Income Support and Labor Rights

Rather than advocate for technophobia, Khan urges a proactive policy agenda to manage the transition. Central to his recommendations is reshaping education systems so they prepare learners for a labor market that will value different combinations of skills. He stresses teaching higher-order thinking, adaptability, lifelong learning pathways and digital fluency rather than focusing narrowly on credentials that A.I. can replicate or augment.

Beyond education, Khan calls for strengthened social safety nets to support workers during periods of displacement. He suggests policies including wage insurance, portable benefits for gig and contract workers, expanded unemployment support tailored to retraining, and incentives for firms that retain and reskill staff. He also highlights the potential role for public investments in sectors where human interaction adds clear value—healthcare, caregiving and education—while acknowledging these areas too will be reshaped by A.I.

Finally, Khan stresses the importance of labor rights and bargaining power, arguing that workers and unions should have a role in shaping how automation is applied within firms so that the benefits of productivity gains are broadly shared.

Debate Over Timing and Scale

Khan’s warnings have triggered debate among economists, technologists and policy makers. Optimists point out that A.I. also creates new kinds of work: roles in A.I. safety, prompt engineering, oversight, domain-specific model tuning, and creative tasks that combine human sensibility with machine augmentation. These voices argue that historically automation raised productivity and living standards, creating new industries and jobs that were hard to foresee beforehand.

Critics of the more alarmist view counter that some tasks will remain human-centric because of social, emotional and ethical complexities. They also note potential for policy measures and private-sector strategies—such as phased deployment, human-in-the-loop systems and stronger retraining programs—to mitigate displacement. Still, Khan’s central contention is about preparedness: even if new jobs emerge, their timing, scale and geographic distribution may not align with the needs of workers who lose employment.

Who Is Most at Risk?

Khan points to several groups likely to be disproportionately affected: mid-skill white-collar workers whose tasks are modular and codifiable; workers in routine administrative and clerical roles; and those in smaller firms or regions without robust retraining infrastructure. He also flags that younger and less experienced workers entering the labor market might find fewer on-ramps to well-paid, secure employment during a period of rapid technological turnover.

At the same time, he warns that focusing only on demographic or sectoral risk can obscure the systemic nature of the challenge: dislocation in one industry can cascade, affecting suppliers, local economies and public finances.

Calls for Collective Action

Underlying Khan’s essay is a call for collective action: governments, educators, employers, labor organizations and civil society should coordinate to shape a transition that is equitable. He urges accelerated public investment in lifelong learning systems, expanded apprenticeship programs that adapt to A.I. workflows, and policy experiments such as wage insurance pilots and regional transition funds to help communities adjust.

Khan also argues for responsible corporate governance, where firms deploying A.I. disclose likely workforce impacts and invest in worker reskilling as part of deployment plans. Such transparency, he suggests, can help craft more constructive transitions and reduce social friction.

Implications for Education Entrepreneurs and Institutions

As the founder of a large educational platform, Khan’s perspective is informed by experience building learning tools and seeing how technology can both expand access and alter pedagogical priorities. He proposes that schools, universities and training providers expand modular, competency-based offerings that recognize incremental skill acquisition and allow workers to upskill while employed.

Khan also emphasizes using A.I. as an educational tool—personalized tutors, automated feedback systems and scalable content creation—while maintaining human mentorship and assessment to cultivate judgment, ethics and interpersonal skills that are harder to automate.

Conclusion: A Test of Civic and Policy Capacity

Sal Khan’s essay frames the rise of advanced A.I. as a test of civic institutions and public policy. He acknowledges the enormous potential benefits of A.I.—productivity gains, better services and expanded access to information—but warns that without deliberate planning and investment, the social and economic costs could be concentrated and severe.

Whether governments and societies respond with the urgency Khan urges will likely shape economic outcomes and social cohesion in the decades to come. His message is both a warning and an invitation: to imagine new education systems, social supports and workplace norms that distribute A.I.’s benefits while protecting those most at risk from rapid technological change.