Part two of three. Visit here for part one.
In part one of this series, I argued that the AI learner is already here. They are already in our classrooms, both physical and virtual, and already using AI as a matter of routine. A contrast to many institutions that are still designing experiences for a learner who showed up a decade ago.
So, as the AI learner solidifies their new normal of learning, this begs a potent question: what is education actually for, when AI can produce a passable answer to most of what we've historically asked learners to do?
The traditional higher education value proposition
For most of the last century, higher education built its value on three pillars, each one resting on a kind of scarcity.
- Curated content. Faculty held expertise learners couldn't easily access anywhere else. The classroom was the sole destination for learning to be shared, and often the only realistic way to get it.
- Sequenced delivery. Curricula laid out material in a structured, expert-crafted order because constructing that path on your own was genuinely hard.
- Credentialed proof. A diploma or degree certified completion of that knowledge transfer, and employers trusted it because there wasn't a better signal for distinguishing talent.
- The model worked because the scarcity was real, and expertise was hard to access. Learning scaffolding was hard to build on your own. Credentials were hard to earn.
And, then came AI.
That doesn't make education less valuable. However, AI makes the source of its value different and the model in which learners choose to determine its value.
What employers are actually saying
When it comes to understanding what a modern, AI-centered education should deliver, employers are the most meaningful voice in the room. The World Economic Forum's Future of Jobs Report 2025 surveyed more than 1,000 employers representing 14 million workers across 55 economies. Their finding is striking: 39% of core skills will change by 2030. That's not a gradual shift. That's a significant disruption to the baseline of what it means to be prepared.
The skills employers say will matter most between now and 2030, in order:
- AI and big data (87%)
- Creative thinking (73%)
- Analytical thinking (70%)
- Technological literacy (68%)
- Resilience, flexibility, and agility (67%)
- Curiosity and lifelong learning (67%)
- Leadership and influence (60%)
AI and big data lead the list, which makes sense. But the rest of it is certainly worth merit. It isn't the content of any particular discipline. It isn't a body of facts. It's human capabilities—the ones that apply knowledge to complex, ambiguous situations. Creativity, analytical thinking, resilience, curiosity, judgment under pressure.
These are the capabilities AI doesn't replace. And they're the capabilities employees need to actually leverage and manage AI well.
The shift that follows
In this context, the value of education from the AI learner shifts from delivering knowledge to developing capability. When AI can produce a passable answer, the work of education becomes producing a person who can tell whether the answer is right, knows what to do with it, and can keep growing as the right answer changes.
That reframing touches everything downstream, and most importantly, changes what we assess and how. It also changes what counts as evidence of learning, and how we think about a degree. Instead of it being the end of a knowledge transfer, it’s positioned as one stage in a much longer arc of capability development, and one that’s modern and in line with how the AI learner sees education.
But, it also changes what we ask of educators. Contrary to what many may believe, faculty aren't being displaced by AI. They're being asked to do a harder version of what they've always done: shape the judgment of a person, not just transfer the knowledge of a field. That's a more demanding role, and a more meaningful one. Not easy, but as we continue to understand the AI learner, educators can build curricula that adhere to the new model of learning.
In part three, I'll get into what designing for capability actually looks like in practice; three capabilities the AI learner needs to build, three design principles to support them, and the institutional shifts required to make it real.
