The University, the Chatbot, and a Call for a New Mission for Higher Education

For centuries, the higher education system has had two central functions: transmitting knowledge and preparing the next generation for professional life. In an information economy where a well-stocked human mind was the ultimate asset, this model generally served us well. Today, higher education is reeling from a collapse in public trust, an increase in federal interference and control, and a daunting mix of demographic shifts, cost pressures, and unsustainable business models. And now, artificial intelligence (AI) is challenging the monopoly that higher education holds on expertise and is reshaping the social contract between knowledge, work, and human value. In an age when information is overwhelmingly abundant and AI can perform most of the cognitive heavy lifting—from computation to writing—more efficiently than any person, the transmission of knowledge is no longer the defining advantage of colleges and universities; it is their greatest vulnerability if it remains the central mission of their work.

This shift does not diminish the role of higher education; it simply renders its old purpose insufficient. When expertise is instantly available on demand and a chatbot can, for example, deliver a personalized explanation of Kant's philosophy at three in the morning, the value of paying for class time to learn from an expert professor is greatly diminished. If institutions continue to focus solely on delivering information and policing student cheating, they will fail the students they aim to serve. The challenge now is not to compete with machine cognition but to redefine human value and the ways we work with the machine. We must equip students for their emerging roles as stewards of intelligent systems. This means preparing them to integrate vast knowledge across domains, align technology with human values, and navigate a world built alongside AI. In short, higher education needs a new contract with society.

The Human Crisis Beneath the Cognitive Revolution

This cognitive reset is unfolding against a disquieting background: a generation marked by record levels of anxiety, depression, and loneliness. These challenges threaten to undermine the very adaptability and resilience the modern economy demands.

Students are entering a workforce in which they must constantly retool and reskill. In a recent keynote address, Matt Sigelman of the Burning Glass Institute noted that approximately 37 percent of the skills required in the average occupation change within five years, underscoring the urgency for more agile, continuously updated curricula.^Footnote1 Sigelman emphasized that the future of work depends on blending skills across domains, as high-value roles increasingly integrate data literacy, computational thinking, and business problem framing. For example, data scientists need not only mathematical and programming expertise but also the commercial insight to translate analytics into strategy and the interpersonal skills to work with cross-functional teams. As routine cognitive labor is automated, the human advantage moves toward the messy, complex capacities: metacognition, ethical reasoning, collaboration, and emotional regulation. Experts at the World Economic Forum and the Organisation for Economic Co-operation and Development (OECD) consistently point to these "human skills" as the primary drivers of performance in uncertain environments.^Footnote2

However, employers report significant gaps in these same capacities, citing shortages in adaptability, collaboration, emotional intelligence, and ethical judgment among graduates.^Footnote3 While AI can approximate such abilities in narrow contexts, it lacks the embodied awareness, contextual understanding, and moral discernment that define human intelligence in real time. The edge is no longer in what we know but in how we think, connect, and act. Many educators and employers would likely attest to the erosion of soft skills in a generation that came of age online, shaped by social media and the long isolation of a pandemic.

To cultivate the human skills that set people apart from machines, educators must first create the conditions necessary for their development. Learning is fundamentally state-dependent. Students arrive at colleges and universities more disconnected than ever from the basic practices that sustain both learning and life: adequate sleep, purposeful movement, social connection, and a sense of mission. Neuroscience confirms that physical and mental well-being are not peripheral luxury items; they are the physiological infrastructure for attention, memory, and judgment. Research has shown that exercise is as effective as antidepressant medication or psychotherapy in treating mild to moderate depression and anxiety, while also enhancing neuroplasticity and cognitive performance.^Footnote4 Adequate sleep is essential for memory consolidation and creativity. High-quality social connection is a powerful health determinant, stress buffer, and predictor of persistence. These are not passing wellness fads; they are the underpinnings of human functioning. When the body and mind are dysregulated, learning is impaired. When education neglects the fundamental biological and relational conditions of flourishing, it undermines its own cognitive aims.

Well-Being as a Design Opportunity

On many higher education campuses, mental health is treated like an emergency room problem, siloed in an overworked counseling center or farmed out to a network of outside providers. However, some of the most effective mental health strategies—regular exercise, structured mindfulness, and better sleep hygiene—are simple, healthy lifestyle practices. Residential colleges and universities often provide exercise facilities and offer programs to support mental health and connectedness, but these needs are often considered as being outside or supplemental to learning and, therefore, easily ignored. They should instead be treated as integral parts of a student's learning journey, designed with intention, and measured for efficacy.

Residential colleges and universities already possess the structure and levers that shape daily behavior: class schedules, classroom norms, assignments, and institutional culture. When institutions intentionally design these elements to embed and reinforce healthy behaviors—making them part of the academic rhythm rather than optional add-ons—adherence increases dramatically.^Footnote5 Colleges and universities have not only an opportunity but a real need to evolve from intellectual ecosystems to behavioral ones. The path to the higher cognition we seek to cultivate must be grounded in foundational acts of self-stewardship.

From Learning Platform to Human Development Platform

If the job of higher education is to prepare students for the society they will enter—one that is globally connected, politically polarized, and technologically accelerated—then its mandate must expand. Career preparation alone is no longer enough. The goal must not only be creating employable graduates but also forging citizens who can exercise sound judgment, collaborate across differences, and be stewards of complex systems. In a world that is flooded with disinformation, drifting toward authoritarianism, and fragmented by social media echo chambers and factionalism, higher education must reclaim its long-abandoned mission of cultivating good citizens. Institutions must cultivate graduates who are equipped with empathy, judgment, collaboration, and—dare we say it—moral reasoning.

The Matter and Space team was developing what we first called a new learning platform but we now refer to as a human development platform. This platform, dubbed LE-1, is now being used at Southern New Hampshire University. It attempts to build the conditions for growth, relational intelligence, and well-being directly into the learning process. Decades of empirical evidence indicate that social-emotional learning and well-being interventions improve academic performance, engagement, and retention.^Footnote6 However, these programs are often isolated from the core curriculum. LE-1 seeks to unify them.

The design of the platform draws from behavioral science and precision health. With the student's consent, LE-1 gathers extensive student data to identify optimally teachable moments through multimodal physiological, behavioral, and contextual signals. It delivers timely, evidence-based strategies through AI-powered academic and wellness agents; reflective micro-interventions; and just-in-time prompts that develop and strengthen understanding, focus, and self-regulation. It also analyzes causal impacts at both individual and aggregate levels, generating dynamic playbooks of what works, for whom, and under what conditions.

This feedback loop transforms education into an adaptive ecosystem that supports learning, self-awareness, and healthy practices. Learners not only receive interventions but also can understand and track their cumulative effects over time, seeing how small habits such as exercise, sleep, and reflection shape learning, well-being, and interpersonal skills.

This data-intensive approach raises important questions about privacy, consent, and potential surveillance. The LE-1 system incorporates opt-in consent and transparent data governance. However, institutions adopting such systems must carefully consider how to protect vulnerable students while maximizing the benefits of personalized support. The ethical framework for educational data collection remains an evolving conversation that will require ongoing stakeholder engagement.

The Vexing Question of Connection

AI shows significant promise for mental health support. Large language model (LLM)-based systems can offer twenty-four-hour availability, stigma-free engagement, and infinite patience—qualities traditional therapy cannot match. Recent meta-analyses show that LLMs can deliver measurable improvements in mood, anxiety, and perceived well-being compared with control conditions.^Footnote7 LLM-based "therapy" systems have already shown benefits comparable to human-led interventions for mild symptoms of depression and anxiety, primarily through increased accessibility, immediacy, and anonymity.

Encouraging findings are also emerging from other educational and wellness contexts, where AI agents act as reflective companions, prompting self-awareness, emotional labeling, and cognitive reframing. In our own work, formal assessments show that users often describe interactions with our reflective agents as "grounding" or "motivating."

But beneath the promise lies a moral dilemma. Surveys and qualitative analyses indicate that adolescents and young adults are increasingly using conversational AI to cope with loneliness and relationship problems, and even to seek support for suicidal thoughts.^Footnote8 While some report feeling less alone, others describe confusion about the boundaries between authentic connection and simulation. This phenomenon is not trivial. As loneliness and mental health challenges intensify, this trend raises difficult moral questions: Which forms of connection can machines safely and effectively mediate, and which must be mediated by humans? In the most tragic cases, parents have filed lawsuits after their adolescent children took their own lives, alleging that extended interactions with chatbots deepened their distress rather than alleviating it.^Footnote9

Attempts to build wellness and safety agents within LE-1 have made these tensions impossible to ignore. While the agents can provide gentle check-ins, mindfulness prompts, and reflective dialogue that many users find genuinely supportive, creating systems that respond compassionately to distress without overstepping into therapeutic territory or offering false reassurance remains deeply challenging. Designing such agents requires balancing three imperatives: efficacy, safety, and authenticity. To be effective, they must respond in empirically validated ways. To be safe, they must adhere to strict ethical and legal boundaries. And to be authentic, they must acknowledge their artificiality without undermining the user's sense of being seen and understood. Each of these aims can come into conflict with the others.

Perhaps the most persistent design dilemma is how anthropomorphic design can foster overdependence on a chatbot. Human brains are wired to attribute agency and empathy even to minimal social cues, such as a voice, a turn of phrase, or an attentive pause. When an AI responds with what feels like compassion, users often infer intention where none exists. This creates both opportunity and risk. The same mechanism that can comfort someone can also mislead them about the capacities and limitations of the system. When we designed LE-1, we intentionally made the boundary between human and machine visible. The agents are warm yet transparent, and helpful yet humble, repeatedly reminding the learner that genuine healing and growth require human connection.

The imperative is clear: Our task is not to make machines more human but to use them in ways to support human wholeness.

The Central Human Project of the 21st Century

In the age of AI, the humbling reality is that humans are often no longer the smartest agents in the room. This shift can feel threatening, especially as AI begins to displace people from well-paying jobs. But it is also an invitation. Freed from the grinding labor of information processing, people can redirect their finite time and attention toward the deeper work of wisdom, connection, and meaning.

Higher education now faces a choice: Cling to the fading mandate of knowledge transmission and compete with machines in their own domain or rise to a higher calling to become the primary civic institution for cultivating whole human beings capable not only of thinking clearly but also of living well within complex, interdependent systems. The global mental health crisis, rising populism, the demonization of others, and growing geopolitical conflict signal an urgent need for a new kind of higher education.

This is not the first time American higher education has been reimagined in response to social and technological change. Recognizing that the U.S. higher education system was unequal and ill-suited to the needs of a rapidly industrializing economy, Abraham Lincoln signed the Morrill Act of 1862, which helped establish land-grant colleges focused on practical education. If the Morrill Act changed the how of higher education, the GI Bill changed the who, expanding access to millions. Today, the U.S. higher education system faces similar challenges. Institutions must reinvent themselves to address the what of higher education, taking a hard look at their role in a fractured world.

To lead in the age of AI, colleges and universities must pursue guided integration, uniting intellectual rigor with emotional intelligence, disciplinary expertise with moral discernment, and personal growth with social responsibility. Campuses must become behavioral ecosystems that promote rest, reflection, and purpose. Curricula must cultivate self-awareness and systems thinking alongside technical skills.

Since the 1960s, higher education has steadily moved away from emphasizing shared values, moral direction, and intentional character development. These areas are now treated as too politically charged to address. While some institutions—including HBCUs, faith-based colleges and universities, and military academies—maintain this tradition, even they must broaden their missions to support mental health and well-being, foster self-development, and teach systems thinking alongside cognitive expertise.

The central project of this century may not be technical mastery but moral and systemic maturity: a collective effort to learn how to live and flourish within an interdependent web of human, ecological, and artificial intelligences. Education holds the power to guide this transformation, shaping not only how people think but also how they coexist, create, and care in the age of intelligent machines.

Notes

1. Mark Sigelman,"Keynote: The Future of AI and the Workforce," posted October 13, 2025, GRAILE AI, YouTube video, 53:09.Jump back to footnote 1 in the text.↩
2. Organization for Economic Co-operation and Development,"OECD Learning Compass 2030," 2023; World Economic Forum, The Future of Jobs Report 2023, (World Economic Forum, April 2023).Jump back to footnote 2 in the text.↩
3. Ibid. Jump back to footnote 3 in the text.↩
4. Ioannis D. Morres et al.,"Aerobic Exercise for Adult Patients with Major Depressive Disorder in Mental Health Services: A Systematic Review and Meta-Analysis," Depression and Anxiety 36, no. 1 (2018): 39–53. Jump back to footnote 4 in the text.↩
5. Phillippa Lally et al. "How Are Habits Formed: Modelling Habit Formation in the Real World," European Journal of Social Psychology 40, no. 6 (2009): 998–1009.Jump back to footnote 5 in the text.↩
6. Rebecca D. Taylor et al., "Promoting Positive Youth Development through School-Based Social and Emotional Learning Interventions: A Meta-Analysis of Follow-Up Effects,"Child Development 88, no. 4 (2017): 1156–1171; Joseph A. Durlak et al., "The Impact of Enhancing Students' Social and Emotional Learning: A Meta-Analysis of School-Based Universal Interventions," Child Development 82, no. 1 (2011): 405–432.Jump back to footnote 6 in the text.↩
7. Han Li et al., "Systematic Review and Meta-Analysis of AI-Based Conversational Agents for Promoting Mental Health and Well-Being," npj Digital Medicine 6, no. 236 (2023). Jump back to footnote 7 in the text.↩
8. Ryan K. McBain et al., "Use of Generative AI for Mental Health Advice Among US Adolescents and Young Adults,"JAMA Network Open 8, no. 11 (2025):e2542281; Wojciech Pichowicz, Marek Kotas, and Patryk Piotrowski, "Performance of Mental Health Chatbot Agents in Detecting and Managing Suicidal Ideation,"Scientific Reports 15, no. 31652 (2025). Jump back to footnote 8 in the text.↩
9. Rhitu Chatterjee,"Their Teenage Sons Died by Suicide. Now, They Are Sounding an Alarm About AI Chatbots," NPR, September 19, 2025.Jump back to footnote 9 in the text.↩

---

Tanya Gamby is Vice President of AI Learner Development at Southern New Hampshire University.

David Kil is CEO and Founder of CML Insight.

Rachel Koblic is a Learning Architect and Product Strategist.

Paul LeBlanc is Visiting Scholar and Special Advisor at the Harvard University Graduate School of Education.

Mihnea Moldoveanu is Director of Desautels Centre for Integrative Thinking and Rotman Digital and the Marcel Desautels Professor of Integrative Thinking and Professor of Economic Analysis at University of Toronto.

George Siemens is Chief Artificial Intelligence Officer at Southern New Hampshire University.