Higher education is facing increasing scrutiny around justifiable concerns related to access, affordability, adaptability, preparation for employability, relevance and value. These are not new critiques, but they have become increasingly difficult to ignore as the gap widens between how institutions of higher education are structured and how knowledge is created, accessed and applied in the world beyond them. The prevailing model, with its fixed curricula, time-bound progression and standardized assessment, was developed decades ago for a context that bears little resemblance to today’s realities. Artificial intelligence is accelerating this tension, not just by enhancing learning systems but also by exposing structural limitations on how higher education defines learning, measures achievement and communicates value. The question is no longer whether students can access knowledge, but whether institutions are structured to help them interpret, question and use it. Much of higher education remains organized around assumptions that are increasingly misaligned with how learning occurs or capability is demonstrated.
From Knowledge Scarcity to Intelligence Abundance
The modern university was built in a world where knowledge was scarce, expensive and difficult to access. Lectures, curricula and degrees emerged as efficient mechanisms for organizing and distributing that scarce resource. A single expert could transmit knowledge to many learners with institutions serving as both gatekeepers and validators. Adaptive and online resources catalyzed a change. Appropriately trained AI systems can now explain concepts in multiple ways, generate examples at scale, simulate complex scenarios and provide continuous feedback tailored to the learner. These advancements shift the operative constraint from a learner’s access to information, to the ability to interpret, evaluate, synthesize, integrate and use it. Such an evolution not only creates abundance, destroying the myths of scarcity, but also moves the focus to demonstrating competency.
Unbundling the Traditional Value Proposition
Historically, universities have developed a system around four tightly coupled aspects: access to knowledge; access to those who create and disseminate it; access to networks of peers, alumni and potential employers; and access to credentials that signal capability. The credential, the degree or certificate provided a signal (or proof) to society—most importantly to a potential employer—that the student had completed a standardized program of study, could reasonably follow instructions, could think within predefined bounds, could last through a process (generally four years) and could be reliably expected to produce value. That coupling is now breaking down.
Knowledge and its creators are now broadly accessible. Networks are increasingly global and digitally mediated, and credentials, once the primary signal of capability, are being supplemented and, in some cases, challenged by demonstratable skills, portfolios, competencies and real-world performance. In engineering, this shift is particularly evident. The ability to generate code, simulate systems through sets of equations or produce design alternatives, is no longer limited to those with deep technical training. AI can assist novices with these tasks. What differentiates individuals is no longer their ability to execute routine processes but their ability to define problems, apply appropriate constraints and evaluate solutions based on multiple complex objective functions, some of which may well have a high level of ambiguity and even conflict. The four-year degree, long treated as a proxy for capability, is an increasingly weak signal. This change does not, however, diminish expertise but elevates it, placing a premium on judgment, context and integration.
Reframing Learning From Delivery to Design
If AI can deliver content more efficiently and more personally than traditional lectures, then the value of the university cannot remain anchored in content delivery alone. Instead, the institution’s role must shift to deliberately designing learning experiences. In engineering education, this practice is a natural extension of long-standing pedagogical approaches such as design studios, laboratories, project-based learning and internships. AI enables the ability to scale and personalize experiences in ways previously not possible.
AI-driven systems can identify where a learner is struggling, generate alternative explanations or examples, create tailored practice scenarios and provide immediate feedback without penalty or judgment. This level of personalization has important implications for learning and acquiring competencies, since a significant barrier to learning is not cognitive but emotional, related to the fear of failure, leading to hesitation to ask questions and a reluctance for students to engage when uncertain. AI-mediated learning environments can reduce these barriers, allowing students to persist until they have achieved mastery. Importantly, this AI usage does not diminish faculty’s role but elevates it, shifting from broadcaster of information (sage on the stage) to designer of meaningful learning experiences, curator of problems and projects, mentor of intellectual development and guardian of academic and professional standards.
Assessment in an Era of Generative Capability
For decades, higher education has been predicated on a specific concept of learning which followed a preset formula defined by the syllabus, standard rubrics and standardized assessment. The model rewarded the ability to reproduce knowledge and solve problems with known solutions, focusing on the “what” and “how to” rather than the “why” and “what if,” emphasizing compliance over curiosity and knowledge of theory rather than its application to new aspects. At a time when AI can generate polished output, the output itself is no longer sufficient evidence of learning, necessitating a shift toward assessment models that emphasize reasoning and decision making, problem formulation and framing, application under constraints and in dynamic and uncertain environments.
In engineering, this shift allies closely with professional practice. Engineers are evaluated not only by the correctness of their calculations but by their ability to interpret requirements, navigate trade-offs and deliver solutions under real-world conditions. In an AI-enabled world, assessment must move from episodic, high-stakes evaluation to continuous, embedded validation of capability, focusing on demonstration of competency under challenging simulated conditions that require systems-level thinking and integration of skills and knowledge.
Decoupling Time From Competency and the Rise of Learning in Work
One of the most consequential implications of AI-enabled learning is the separation of time in seat from the achievement of competency. When learners have access to continuous guidance, immediate feedback and unlimited practice opportunities, the traditional linkage between time and mastery, which is itself artificial, becomes increasingly tenuous. The pace of change in the workplace is continuing to accelerate to the point where skills acquired early in a student’s period of undergraduate study may well be outdated before graduation, creating a fundamental misalignment between education and application.
This acceleration not only necessitates greater adaptability and continuous, rapid updating of curricula to ensure relevance but also emphasizes the criticality of better integration of learning and work experience. In engineering, cooperative education and internships have long demonstrated the value of experiential learning. AI extends this value by enabling complex environment simulations, scenario-based practice at scale and integration of real-world situations into learning tasks. The result is a model in which learning is not confined to discrete academic activities but embedded within professional activity. Education becomes a continuous process of capability development rather than a separate phase preceding employment.
From Generalists to AI-Enabled Domain Experts
A common concern is that AI will dramatically decrease the need for human expertise. It is more likely to increase the value of deep domain knowledge. As AI lowers the barrier to executing routine and even moderately complex tasks, the critical differentiator becomes the ability to understand context, define meaningful problems and interpret results. This differentiator applies across disciplines, from engineering and science to business and the social sciences. In engineering, for example, AI can generate multiple design options. It cannot, however, without expert guidance, determine which design best satisfies multiple objective functions including safety requirements, regulatory constraints, sustainability goals and economic considerations. Those judgments require expertise. This distinction is important.
The competitive advantage shifts from execution to problem set-up and judgment. The ability to ask the right question, define the correct problem and select among competing solutions becomes more valuable than the ability to execute any single method. The future does not belong to those who simply know how to use AI tools. It belongs to those who can combine deep domain knowledge and expertise with AI capabilities to achieve outcomes that neither could produce alone. In this sense, higher education’s goal is not to produce AI specialists in isolation but to produce AI-enhanced experts—individuals who understand their fields deeply and can leverage AI as a force multiplier.
Extending Learning From Isolated Innovations Toward Intelligent Learning Systems
With AI, access relates to ability to gain knowledge and intelligent guidance. Personalization evolves into fully adaptive learning pathways, scalability extends from enrollment to individualized support at scale, assessment shifts from episodic evaluation to continuous validation of capability, content creation becomes dynamic with materials generated and updated in real time, and workforce alignment tightens as learning integrates directly with professional practice. While each of these shifts is significant, viewing them independently risks underestimating the collective impact. Taken together, they redefine the architecture of higher education, moving it from the traditional model organized around constraints of time, location and standardization, to one focused on what a learner can understand, apply and demonstrate, regardless of when or where that learning took place.
We are fundamentally moving from an institution-centered model of education to a learner-centered system of capability development, best understood as the emergence of intelligent learning systems in which learning pathways adjust dynamically with continuously evolving content and experiences, feedback is immediate and contextual, and capabilities are developed through continuous interaction between the learner, the problem space and increasingly capable technological systems.
Amid these changes it is important to recognize that institutions of higher education remain uniquely positioned to provide environments for deep intellectual engagement, opportunities for collaboration and community, spaces for identity formation and professional socialization, and contexts for ethical reasoning and judgment. These are not peripheral functions. They are central to the development of individuals who can use powerful technologies responsibly. The risk is not that AI diminishes the university. The risk is that institutions remain bound by models built around assumptions of value that may no longer be valid, such as adding new buildings and maintaining programs with growing disconnect from contemporary relevance and value to society and the communities they serve.
A Strategic Path Forward
The challenge before higher education is not simply the adoption of AI tools but the re-envisioning of learning systems that include them. Institutions that succeed will be those that move from fixed curricula to adaptive learning pathways, from time-based progression to demonstrated capability, from episodic assessment to continuous validation, and from content delivery to design of experiences. The future of education will not be determined by whether AI is used but by how thoughtfully it is integrated into the structures that shape learning. The question that all leaders must consider is whether they will move their institutions to greater relevance and value, ensuring a prominent place for institutions of higher education in the future, or they will allow for systematic pre-truncation of options because of perceived and real constraints and boundaries set decades ago, becoming increasingly disconnected from the learners and the communities they are meant to serve.