Professor Mitchell Stevens, Stanford University: Linking the Conversations About AI, Learning and Longevity

Van Ton-Quinlivan
Hello, I’m Van Ton-Quinlivan, CEO of Futuro Health, welcoming you to WorkforceRx, where I interview leaders and innovators for insights into creating a future-ready workforce.

The age of AI has become a common way to refer to the times we live in, but you could more accurately call it the age of AI anxiety because recent national surveys show more than half of Americans are worried that AI will replace a significant share of their jobs.

But of course, this isn’t the first time technological leaps have disrupted our economic lives. And our guest today, organizational sociologist Mitchell Stevens, says the lesson to draw from previous responses to these challenges is to invest in people and their education.

Mitchell pursues his longstanding interest in lifelong learning and related topics as a professor in the Graduate School of Education at Stanford University, where I am also an alum. He is also a faculty affiliate of the Institute for Human Centered Artificial Intelligence and helps lead the Center for Longevity. Most recently, he’s convened the Center’s Learning Society Initiative in which I’m participating.

Thanks very much for joining us today, Professor Stevens.

Mitchell Stevens
My pleasure, Van. Please call me Mitchell.

Van
Absolutely. Well, let’s begin. Should people be anxious about AI’s use in the workplace and displacement of jobs?

Mitchell
I would say “yes, and.” The reason they might be anxious is because the sense that change is happening as we speak to every aspect of our work and non-work lives is very real, but I think any good therapist would say that anxiety by itself only gets one so far. So, I do think there are other emotions with which to face this moment.

One is curiosity, a desire to think through an experiment with just all the different ways in which these new technologies might change our lives for the better. And also a determination, which is what I’m really not seeing in the country at present, to make this current moment of historical change a strong net positive for our people over the next two to ten years.

Van
As a sociologist, tell us about the lessons you’ve drawn from previous periods of technological disruption and how Americans have responded to them.

Mitchell
Well, one of the great uses of economic history is we can remind ourselves that while AI itself is new, this is hardly the first time that major technological change has resulted in a big transformation of the country.

I point in my work — and also in the Learning Society project — to the so-called second industrial revolution. That’s the period that spans roughly between the Civil War and World War I when industrialized countries all over the world figured out how to harness power for mass production and utility. That’s when we created regional electrical grids. That’s when we harnessed steam and fossil fuels to run factories.

That industrial revolution reshaped virtually every aspect of American life. One of the ways it did that was by encouraging us to create universal mass schooling, a recognition that in order to enable people to flourish and prosper in this new world, new sets of skills were required of everyday people, as basic literacy and numeracy among them.

So one result of the industrial revolution was a massive investment in basic skill development for everyday Americans. And while we didn’t do that uniformly well, I would call that a substantial step forward in terms of progress and empowerment of everyday people.

Van
So based on that, to help with our AI anxiety, what lessons can we borrow from those responses and what else should we be adding to develop a national vision for human capital investment?

Mitchell
Well, what I find troubling, Van, both nationally and even here in Silicon Valley — which is certainly not known for its pessimism…this is a part of the world where we tend to see technology in optimistic and net positive ways – but not so with AI, I fear. It’s often being framed by academic experts, political leaders, and many cultural commentators as something that’s happening to us, right? It’s a steamroller that’s coming. We have to somehow defend our lives, defend our children, defend our institutions from this technological change.

There are two large omissions there. AI is not a thing independent of organizations, many of them large corporate firms. AI doesn’t make decisions for itself. People and organizations make decisions about how AI will be deployed. And the other is we’ve somehow forgotten that technology is supposed to enable and catalyze human flourishing. It’s not supposed to threaten human flourishing. Somehow we haven’t yet turned that corner.

We haven’t recognized that the reasonable anxiety that we face in the wake of these transformations also creates extraordinary opportunities for us to fundamentally rethink how work takes place, how divisions of labor most appropriately unfold, how work should fit into the arc of ever-lengthening lives.

I’m not trying to say that all of the future will be all good, but only that the future can only be good if everyday people, politicians, corporate leaders, sort of ambitiously anticipate a positive future and then imagine ways to build to make that future come to be.

Van
So Mitchell, you have been inviting us to come together and think more broadly and kind of pull out a little bit and imagine this positive future. Tell us more about the Learning Society Initiative and how you think it will unfold.

Mitchell
My pleasure. This is a conversation that began at Stanford in the fall of 2024 under the auspices of two organizations, the Center for Advanced Study in the Behavioral Sciences and the Stanford Center on Longevity. I want to talk a minute about the second center specifically.

First of all, Van, you’ll note that it is not called the Stanford Center on Aging. There is a national and international conversation about aging which sort of mimics the problem that I’m suggesting here. When I say aging, people tend to think frowny face, right? As in anti-aging cream, right? Or people worry a lot about the costs of a so-called aging society that that’s gonna take more resources for us to take care of our elderly.

Why not have a conversation about longevity instead? Longevity is a positive change of the 20th century. In fact, it’s partly a positive change of the industrial revolution I mentioned earlier. The director of the Stanford Center on Longevity, Laura Carstensen, likes to point out that human lifespans grew in the 20th century by a greater amount than in all previous human history put together. We added about 30 years to the lifespan between 1900 and the year 2000. That’s an enormous gift of the 20th century.

What have societies done with that? Most of the time they’ve parked it in old age. They’ve made the most costly, least fun part of the life course longer, right? Instead of extending what longevity researchers call functional adulthood, that means the span of the life course in which my chronological age (the number on my birth certificate) my physical age, (my health and well-being), and my working age (my ability to meaningfully participate in work and leisure), have all lengthened.

And we know how to do that. We know how to make functional adulthood longer. Health and nutrition are huge parts of that, certainly adequate medical care. Also education and learning are extraordinarily important as anti-aging mechanisms — not only the quality and amount of education that people receive in the first two decades of life, the so-called Q1 or first quarter of life, — but the education and learning opportunities they have in the second and third quarters of their lives also are very powerful mechanisms of enabling functional adulthood over a longer period of time.

So turning the bad news of aging into the good news of longevity is not a project of denial or putting lipstick on a pig. It’s a project of sort of rethinking what the assets of this new reality are and how to sustain them.

So, the goal of the project that you mentioned that now flies under the banner of Learning Society was to take global conversations about longevity and tie them to global conversations about AI and the future of work, and to recognize, for example, that the fact of declining fertility, fewer children and longer lives means that a lot of the revolutionary changes that are currently underway in the economy are going to be experienced and managed by people in their 30s and 40s and 50s and 60s.

It’s also the case that people in their 30s and 40s and 50s and 60s are going to need to be working longer in order to sustain their own economic well-being, but also their psychological health and enjoyment of life.

What longevity did to aging, I’m hoping learning can do to AI, which is transform the anxiety or at least replace some of the anxiety that comes with AI to a sense of optimism and determination to use these tools to make individual and collective lives better.

Van
So Mitchell, in speaking with Eva Sage-Gavin, who is an advisor on your Center on Longevity and a prior podcast guest, she mentions that you have been having conversations with employers as well as workforce leaders and sharing some of this vision around longevity and the Learning Society. I wonder if you can give us some thoughts that you’re hearing back as part of those conversations.?

Mitchell
Absolutely. What Van and I are calling the Learning Society is the current change thesis that listeners can learn more about at learningsociety.io

The premise of that conversation — which is very much ongoing and anchored at Stanford University, but truly national in scope — is that the United States invested in people in the 20th century by building schools and extending school age to an ever larger portion of the lifespan. That’s how we built what economists call ‘human capital’ in the 20th century. We gave people education and skills by creating universal educational organizations called schools and obliging people to spend more and more time in them. That was a very, very positive way of investing in people, but it had its downsides, what economists call negative externalities.

One of the things we did by putting learning in school was we separated the official site of learning from paid work. We encouraged people to imagine that they either were investing in education or they were investing in paid work, but they somehow couldn’t do both at the same time and in the same place.

The other thing that we did was we ended up stratifying and sorting people on the basis of the school credentials that they had, not necessarily on the capacities and skills that they actually possess. Once you realize that, of course, people can have a lot of schooling but not learn very much, and they can have a lot of knowledge and wisdom even if they haven’t gone to school, you recognize this problem that we did a lot of schooling, but did not necessarily match learning with schooling, nor did we adequately recognize how much learning and how much wisdom and how much capacity could be accrued outside of school.

We think that this is a historic moment to recognize that schooling and learning are not the same and that we can recognize and reward learning and the providers of learning wherever it occurs. So that’s where this question of corporate America comes in. Turns out people learn a lot in the course of doing their jobs, and again, that’s just common sense wisdom, right? You don’t need to have a PhD in education to recognize that the act of doing work is itself an educative experience. We’ve always known that and employers have always known that, but because our society focused so much on schools and school credentials, we ended up discounting all of the learning opportunities and all of the human learning itself that takes place in, for example, workplaces.

Once we’ve set aside the presumption that school is the best or only way to learn something, you can sort of recognize just how much learning opportunity is provided by employers. And what would it mean to sort of systematically recognize employers’ provision of that learning? What would it mean to empower every rank and file employees to realize that they are accruing skills and capacities in the course of doing their work that might be translated into their forward careers.

The insight that learning happens at work is not new, but the idea that we might want to systematically observe, recognize, incentivize, reward, and celebrate that learning an almost revolutionary idea in a schooled society context. So, that’s why we’re spending so much time talking with chief human resource officers and chief learning officers to get their take on — and to empower them also — to help the rest of us sort of think about what it means to be a learning workplace.

Van
And for the education leaders that may be listening in, what is your advice to them?

Mitchell
That’s a really good question, Van. I’m in the School of Education at Stanford, and we spend a great deal of time studying K-12 schools and K-12 educators. I myself study adult learning and higher education. One of the interesting patterns that we’ve seen in terms of receptivity to AI technologies among educators is that in general, K-12 educators have been much more receptive and embracing of the potential of AI tools to enhance and personalize learning for specific children, right? K-12 teachers are often miracle workers doing a lot with a little. They are doing a great deal with AI in terms of specifying lesson plans, customizing academic trajectories, learning from their colleagues across the country in distributed learning communities.

With respect to anxiety, you’re getting a lot more of that on the post-secondary side, and I think that’s the case for two reasons at least. One is, you know, I’m an academic. I earned a PhD. My sense of my knowledge and wisdom as a kind of hard-earned accomplishment is very important to my sense of self. If you don’t need a fancy professor to tell you something because Google AI will not only give you the answer, but summarize the text and also 30 years of critical analysis of that text, why do you need an English professor? And also, if learning can happen all over the place and doesn’t require a college degree, then the primary thing that I’m selling — advanced education, as we used to call it — sort of diminishes in value.

So I think perhaps — incorrectly, but understandably — that universities, academics and academic leaders experience AI as a real threat to their professional expertise and also to their special purchase as educators. But whether we like it or not, the future of knowledge production and dissemination is just going to be different, so it’s our responsibility and opportunity to think through the future of our own profession.

Van
As you know, Mitchell, our focus at Futuro Health is on the future of care and addressing the critical shortage in caregivers, which is only gonna grow as the country ages and doubles its population that is over 65. And as our listeners know, we all need more care as we grow older. It also seems like the caregiving dimension and the skill sets to care for others, as well as being given care, is another phase of learning that all of us who hit a certain age have to attend to. I wonder how we should think about that in context of this lifelong learning?

Mitchell
Van, I’m so glad you raised that. You and Futuro are at the cutting edges of two intersecting transformations — and I’m sure you know this, but it’s become ever clearer to me in the course of this work. Transformation number one is the reorganization of medical expertise. What’s happening to college professors is also happening to radiologists and the people who diagnose health and illness. That expertise will continue to be valuable and important as embodied in human beings, but it’s going to be accessible in very different ways. It already is because of artificial intelligence.

Dr. Google, for example, is very close at hand. So expertise itself is shifting. Expertise is going to become in some ways more democratic and more accessible, right? But care is a different matter. Care is the fundamentally interpersonal, nuanced, contingent, evolving, reciprocal knowledge that one person has for another. Machines are already doing care in ways that they did not 10 years ago, but I am in the camp that interpersonal human care is at some level non-transferable, nor would I want it to be transferable. I would prefer that my mother be a human being rather than a machine. Even though I didn’t have the most perfect mother in the world, I would prefer that that be a human relationship rather than a machine relationship.

The other thing is — as work changes and a lot of the technical analytic capacities that we use our brains for can be transferred to machines — then care, which is in my view, a fundamentally embodied practice, becomes more available to me, right? I may need fewer doctors, but I need way more nurses and allied caregivers.

Or in education, a teacher may not have to worry about preparing her lesson plans and grading assignments, but she can spend way more time interacting with the human beings in her classroom, creating the relationships and emotional environments that she wants to help nurture the young people in her purview.

So on the one hand, this transformation displaces a lot of the “brain work” that took a lot of time and enables us to do the “body work” of care. On the other hand, it enables us to emphasize the value of that work as non-transferable to machines. And is that a complicated, difficult, politically fraught process in medicine as much as in academia? Absolutely. Is it a huge opportunity to address what social scientists call the care deficit in our country? Absolutely.

That’s another way in which I see this as a challenge…the possibilities for a more nurturing country are just huge if we act together and thoughtfully.

Van
Well, I really appreciate the sense-making you just offered there to help us think more broadly about the situation that we find ourselves in and the multiple futures that we could create.

Mitchell
The multiple futures and the futures that we create…that’s up to us, right? We get to decide not how the future will unfold, but we get to decide what futures we want to build. And again, that’s what I want this Learning Society work to do, is to remind people that we get to imagine the possible futures of our lives and our society in the wake of this change.

Van
Well, let me close by offering you the opportunity to answer this question: what makes you optimistic about developing human capital in the age of AI and AI anxiety?

Mitchell
That’s a great question, Van. One of the wonderful things about this moment, I think, is that what our project calls the civic architectures that we built for education, learning, care, the way in which we organize the rhythms of work and the relationship between work and home are all at hazard for change because of artificial intelligence and because of longer lives. The fact of longer lives and the fact of major technological change means that we would be foolish to presume that the way in which we organized schools, colleges and universities and workplaces in the 1950s and 60s and 70s makes sense for this new world.

And the more people I see that are recognizing that, the more it is possible to sort of create new lines of communication among domains that were previously regarded as separate. K-12 education, post-secondary education, workforce development, corporate learning…for years, we thought of those domains as separate. But if you recognize that all of our institutions are being disrupted, and that we need to think about change to enable longer lives, we get to have conversations across professional domains that we never used to have.

I’m absolutely convinced of that because my relationship with you is one example. I’ve known essentially nothing about allied health and workforce development until we began a conversation with each other years ago. It turns out there’s way more connection between Futuro Health and university academics in fields that are far away from health than there were in the past. So let’s keep talking and build a new professional community that is agnostic about the kind of borders that divided us in the past.

Van
Well, thank you for your leadership and in helping us see the future. Thank you for joining us today, Mitchell.

Mitchell
My pleasure.

Van
I’m Van Ton-Quinlivan with Futuro Health. Thanks for checking out this episode of WorkforceRx. I hope you will join us again as we continue to explore how to create a future focused workforce in the nation.