As artificial intelligence reshapes the workplace, there is a growing consensus that mastering AI skills and data tools guarantees career security. Across campuses, there is an expanding rush to create new AI credentials, and universities are reshaping their curricula to meet the demand, fearing that inaction will make students less competitive as they begin their careers.
This assumption is understandable and maybe even inevitable in the moment—but it is also incomplete.
Technical training provides only a portion of what students need. Human curiosity, judgment and ethical reasoning become indispensable when machines automate routine tasks.
The question is not whether students should learn to work with AI. It is whether they will also learn to think about it.
This is where a liberal arts education becomes critical. It is not a checklist of marketable skills, but a sustained engagement with knowledge across disciplines.
It cultivates capacities that AI cannot replicate: asking difficult questions, navigating complexity, exercising creativity and making ethical judgments—responses that don’t necessarily resolve cleanly in practice.
For that reason, a liberal arts education should stand alongside technical training as an essential foundation for understanding and shaping AI.
AI skills in the liberal arts
Teaching narrowly defined skills to the exclusion of the liberal arts may produce immediate results, but it risks producing graduates whose knowledge quickly becomes obsolete. By contrast, a liberal arts education, especially one grounded in a foundational subject or an interdisciplinary major, equips students to navigate rapid and unpredictable change.
It teaches them to listen, to communicate, to think critically and to adapt. It turns experience into insight and curiosity into action.
Like the liberal arts, AI reflects a form of integration and the most effective models draw connections across vast domains of knowledge. Technical fluency, therefore, should be part of a liberal arts education. But fluency is still not the same as judgment.
The difference lies in how that integration occurs. AI integrates information statistically rather than through deliberation, generating coherence through patterned association rather than meaning through understanding.
Nowhere is this distinction more consequential than in practice. In hospitals, AI systems designed to predict sepsis from patient data have sometimes missed serious cases while producing false positives.
Despite training on large datasets, such systems can overlook contextual cues and clinical judgment, requiring physicians to decide in the moment when to rely on them and when, carefully, to override them. The challenge is not only technical accuracy; it is the exercise of personal responsibility under conditions of uncertainty.
The same holds beyond medicine. A workforce trained only in optimization may struggle in situations defined by ambiguity, competing values and ethical consequence. These are not rare cases—they are the conditions under which many of the most important decisions get made.
The capacities required to navigate them are cultivated through sustained engagement with ideas, disciplines and lived experience that is precisely the work of a liberal arts education.
Asking better questions?
The future of work will reward those who can synthesize information, challenge assumptions and act with ethical clarity. It will also at times reward those who can recognize when clarity itself is limited.
The liberal arts prepare students not just to survive in an AI-driven world, but to shape it. To view education merely as skills training is to underestimate what students will be asked to do. They will not simply use powerful tools, they will need to judge when those tools should be trusted, and when they should not.
This emphasis on judgment begins with a more fundamental capacity, the ability to ask questions. Socrates is remembered for a method of teaching in which carefully crafted questions guide others toward understanding.
By probing assumptions and exposing contradictions, he showed that insight begins not with answers, but with questions. In a limited sense, this approach mirrors how prompt engineers shape inputs to guide artificial intelligence toward useful outputs.
But unlike artificial intelligence, human intelligence is not merely responsive. It is reflective and shaped through questioning that cultivates independent thought and moral reasoning.
AI will continue to improve at generating answers faster, cheaper and at greater scale. But will we continue to improve at asking questions?
That is the central challenge of this moment. If education is reduced to technical training alone, we risk producing graduates who can operate powerful systems but lack the judgment to use them wisely.
The future will not be shaped solely by those who can build or prompt AI. It will be shaped by those who can understand its implications, question its outputs and decide when not to use it.
That is not a technical skill. It is a human one.