Guest Opinion | The brain error in AI tutor for schools and education

||| FROM DAVID STEPHEN |||

The [September 1, 2025] Labor Day was a question of what should become of education, if AI can operate intelligence, to do aspects of productive work? What is the missing piece to learn for the future, in the era of AI? How much potential do AI tutors hold, specifically, against what excellent human instructors already provide? 

The biggest problem in education is the unknown — how human intelligence works. The promise of revolutionizing learning is less about another trend, but what is understood as human intelligence and how it should guide learning.

Learning has subsumed a ton of [guess] interventions over decades, with explorations for improved outcomes. Yet, the quality of knowledge as a transcendent property has eluded many. The barriers to the transformational capacity of education are closely aligned with human intelligence, much more than facilities or instructors’ question. 

It is not even progress, at this point, to assume that artificial intelligence can augment learning, without knowing how human intelligence works. The human intelligence that AI would train is the same human intelligence that would compete with AI for work? What should humans be learning for now [that the subtraction of whatever AI can do or answer, from what humans can do, in several important tasks, leaves the thing a lot less valuable — as an impact of labor]? 

Learning Questions 

How does learning adapt to significant problem-solving? What are the techniques to persist in learning complex stuff — that may hold important solutions? How can learning be designed to match patterns to answer the unsolved? If an individual is learning what AI already knows, how should that learning be structured, to out-compete AI?

Will it ever be possible to personalize learning for everyone? 

AI Tutor 

For now, AI tutors, AI schools, AI education or AI learning are not offering anything beyond what is still possible with human instructors. What education is seeking in this era, that AI tutors can optimize for, is pathway displays for memory and intelligence in the brain, to track, almost in parallel — the processes of understanding, recall, creativity, innovation and expertise, to increase the chances for those. 

Simply, what an AI tutor should solve is what a human instructor cannot yet solve : which is the likelihood to target learning, for navigations in the brain, towards advancing humanity. Already, all AI chatbots can answer questions, with several examples — plus simplicity. Still, it is not like grasping, for all, is now straightforward. This is saying that the problem that exists, continues to linger, even if mitigated by consumer AI. 

Human Intelligence 

Any AI tutor program that can develop or show a model of how human intelligence works, to prospect learning [for relays] in the brain would transform education — this century. Simply, show a concept of how human intelligence works, and use that to tailor lessons, for higher-order results. Already, human instructors can provide regular training, but what would mean difference, for AI, is to shape human intelligence, to withstand the uncertainties that AI holds for the future.

AI schools can explore theoretical neuroscience, developing models for many aspects of learning. Memory, conceptually, is obtained in destinations in the brain, but intelligence [which is the use of memory] are navigations [of information] summaries across memory areas. Rote memorization, for example, is the making of new paths between memory locations, so that the paths are available for direct relays subsequently, resulting in recall. Problem-solving could be the overlay of two memory locations, so that their differences, rather than contrast are made similar, opening up possibilities for answers. 

The opportunity for AI in education is not to do what human instructors can do or to assume it will simply make things easier when the [human intelligence] question remains unanswered and, AI would be able do aspects of several jobs, if that is the learning objective. How does human intelligence work is how to restart the education project — now that AI is ascendant. 

ChatGPT Edu

There is a recent [August 28, 2025] announcement by Indiana University, IU strengthens national leadership in AI innovation with ChatGPT Edu rollout, stating that, “Indiana University is expanding its robust artificial intelligence offerings with OpenAI by providing access to ChatGPT Edu, a version built specifically for higher education. ChatGPT Edu offers the world’s most advanced AI tools for learning, teaching and research.

In providing access to all 120,000 students, faculty and staff, IU will be the second largest ChatGPT Edu rollout of all time for OpenAI, demonstrating IU’s national leadership in higher education innovation and underscoring its commitment to responsibly integrating AI across its campuses while preparing a future-ready workforce.

IU faculty and staff can request institutional access beginning Sept. 2, with student access launching Jan. 1 [2026]. The deployment of ChatGPT Edu is one of many ways IU is integrating AI across the academic experience.

In August, IU launched a new, free GenAI 101 course that serves as a foundational program introducing the IU community to generative AI concepts, applications and responsible-use practices.

Together, GenAI 101, ChatGPT Edu and IU’s expanding suite of AI services ensure that faculty, staff and students are not only equipped with powerful AI tools but also prepared to use them effectively and ethically.”

David Stephen currently does research in conceptual brain science with focus on the electrical and chemical configurators for how they mechanize the human mind with implications for mental health, disorders, neurotechnology, consciousness, learning, artificial intelligence and nurture. He was a visiting scholar in medical entomology at the University of Illinois at Urbana Champaign, IL. He did computer vision research at Rovira i Virgili University, Tarragona.