A September 27 Tokyo seminar brought schools, Suginami education officials, and Teikyo University together to examine how teacher preparation should change as AI takes on more instructional and administrative tasks.
Editorial Note
This article covers a professional seminar organized by Uchida Yoko’s Education Research Institute. It does not represent new Ministry of Education policy, a national teacher-licensing requirement, or an official determination that particular teaching functions should be transferred to artificial intelligence.
New To Education has already covered broader Japanese AI-literacy and teacher-workforce issues. This article intentionally focuses on a narrower search intent: how universities, municipalities, and schools may need to redesign the development of teachers when some traditional tasks can increasingly be supported by AI.
A Tokyo education seminar on September 27 asked a question more difficult than whether teachers should use artificial intelligence.
It asked who will develop the teachers needed for an AI-rich school system—and what those teachers should be uniquely capable of doing.
The in-person seminar, titled roughly “Who Will Train Teachers in the AI Era?”, was held from 2 to 5 p.m. at Uchida Yoko’s Tokyo headquarters. It brought together Suginami education leadership, school principals, Teikyo University representatives, university students, and an education journalist for a three-part discussion about the role and development of teachers.
Bottom Line
The seminar’s premise is not that AI makes teachers unnecessary.
Its premise is that AI may increasingly assist with work such as knowledge transmission and preparation of instructional materials, forcing education systems to identify more precisely what human teachers contribute.
The organizers emphasize functions such as noticing subtle changes in students, receiving feelings students have not yet expressed clearly, and supporting children as they learn to think, choose, and act independently.
What Happened
The first part of the seminar used observations from Teikyo University students who had watched teachers working in Suginami public elementary and middle schools. School principals then discussed what those observations revealed about actual day-to-day teaching.
The second part moved from observation to teacher development, bringing municipal education and university teacher-preparation perspectives together. The final panel considered how universities, local governments, and schools should divide responsibilities and collaborate in supporting what organizers described as “thinking teachers.”
What This Means
Much of the education debate around AI begins with tools.
Can AI create a lesson plan? Can it summarize reading? Can it generate quizzes? Can it provide feedback? Can it answer student questions?
Those questions matter, but they are incomplete.
If technology becomes good at parts of the teacher’s workflow, teacher preparation may need to spend less time treating every traditional task as equally central and more time developing the judgment that technology cannot reliably substitute.
That could include observation, relationship-building, diagnosis, ethical judgment, classroom leadership, motivation, interpretation of student behavior, and deciding when not to use technology.
Who This Affects
Universities that prepare teachers are directly affected because program design may need to change.
Municipal boards of education also matter because professional development does not end when a teacher earns a credential.
Schools are the third part of the system.
Teachers learn extensively through mentoring, collaboration, lesson study, observation, experience, and interaction with school leaders.
The seminar’s three-sector structure—university, municipality, school—implicitly recognizes that no single institution controls the entire development process.
Teacher Preparation Cannot Become Prompt Training
There is a risk that “AI training for teachers” becomes overly technical.
Educators may receive workshops on how to generate worksheets, write prompts, summarize documents, or automate routine tasks.
Those skills can save time.
They are not a complete professional-development strategy.
A teacher also needs to know when an AI output is wrong, biased, inappropriate for a student’s developmental level, disconnected from curriculum goals, or likely to undermine the learning process.
That requires subject knowledge and professional judgment.
In other words, stronger AI can increase the importance of teacher expertise rather than eliminate it.
Human Attention May Become More Valuable
Schools generate enormous amounts of visible information: grades, attendance, assignments, test data, behavior reports, and digital activity.
Students also communicate through less measurable signals.
A child becomes unusually quiet.
A normally organized student stops bringing materials.
A student begins avoiding peers.
Another appears engaged on paper but has stopped taking intellectual risks.
Experienced teachers often notice these patterns before a dashboard identifies them.
AI may eventually help detect some signals.
The ethical and relational decision about what to do next is different.
Students are not simply data points requiring optimization.
They are people whose behavior can have multiple explanations, and educators need to respond with judgment and care.
What This Does Not Mean
The September 27 seminar does not establish a national Japanese policy defining “human-only” teacher tasks.
It also does not prove that AI will reliably take over particular instructional responsibilities.
AI capabilities continue changing, and different schools will have different technology, staffing, policies, student populations, and risk tolerances.
Teacher education should therefore avoid designing the profession around predictions that may quickly become outdated.
The more durable goal is preparing teachers who can evaluate new tools rather than simply follow them.
The Bigger Picture
Japan is already dealing with teacher workload, recruitment pressure, curriculum demands, and changing expectations around technology.
AI enters that environment as both an opportunity and a potential distraction.
If it genuinely removes repetitive work, teachers may gain more time for planning, feedback, student relationships, and collaboration.
If AI simply adds new platforms, monitoring requirements, documentation, and expectations without eliminating old work, workload could increase.
Teacher preparation therefore needs to include organizational questions, not merely technical competence.
Schools should ask which tasks can be simplified, which should remain human-centered, and which should disappear entirely.
What Happens Next
The most important follow-up would be translating these conversations into teacher-development design.
Universities can reconsider field experiences.
Municipalities can rethink induction and professional learning.
Schools can examine mentoring and collaborative planning.
Researchers can study which forms of AI support actually improve teacher effectiveness or reduce workload without weakening learning.
Japan does not need one universal answer immediately.
It needs evidence from practice.
Why This Matters
The question “Will AI replace teachers?” is usually too crude to be useful.
A better question is which parts of teaching are routine, which require expertise, which depend on human relationships, and which tasks should not exist in their current form at all.
The September 27 seminar matters because it moves the conversation from software features to professional identity.
If education systems want teachers who can work intelligently with AI, they will have to reconsider how those teachers are prepared, mentored, and supported.
Key Takeaways
Uchida Yoko held an AI-era teacher-development seminar in Tokyo on September 27.
Participants included Suginami education officials, school principals, Teikyo University representatives, and university students.
The seminar examined which parts of teaching remain distinctly human as AI supports more routine work.
University students’ observations of real school practice were used as part of the discussion.
The event focused on collaboration among universities, municipalities, and schools.
It did not establish new MEXT policy or licensing requirements.
The distinct NTE angle is teacher formation and professional judgment, not generic AI literacy.
Frequently Asked Questions
Was this a Japanese government announcement?
No. It was a seminar organized by Uchida Yoko’s Education Research Institute with participants from education institutions and local leadership.
Did the seminar say AI should replace teachers?
No. Its framing emphasized the continuing importance of human teachers while asking how their roles and development should change.
Why were university students involved?
Teikyo University students had observed teachers working in Suginami schools. Their observations provided a practical starting point for discussion about what teachers actually do.
What makes this different from general AI teacher training?
The seminar was not primarily about learning specific AI tools. It focused on teacher identity, judgment, preparation, and institutional responsibility for developing educators.
Final Thoughts
The profession should not define teachers by the tasks AI can imitate most easily.
Teaching is more than producing content.
It involves judgment about people, timing, relationships, motivation, uncertainty, and responsibility.
As AI grows more capable, teacher education may need to become more human-centered, not less.
That is the deeper question Japan’s September 27 seminar put on the table.
Support New To Education
https://newtoeducation.com/support-us
Related Articles
Japan Continues Teacher Workload Reform as Schools Face Growing Staffing Pressure
Japan Reconsiders Curriculum Overload as Schools Look for More Time to Teach Deeply
Sources
Uchida Yoko — AI-Era Teacher Development Seminar
