This is Hamamoto from TIMEWELL.
"Abilities AI cannot replace." It is a phrase often quoted in commentary on the N-E.X.T. High School Vision, but read the Grand Design itself and you find its substance listed concretely: language ability, information-use ability, problem-finding and problem-solving ability, and the ability to collaborate with others[^1].
In this article, I dig into Perspective 1 of the Grand Design that MEXT released on February 13, 2026 (the full overview is here) — "as sovereign citizens who live independently in an uncertain age, nurturing abilities and individuality that AI cannot replace." Even seen from my position, whose core business is supporting AI adoption at companies, the view of ability written here matches, to a startling degree, what is happening right now in corporate hiring and talent development.
Neither "shut AI out" nor "AI can do anything": a two-stage design
The Grand Design's view of AI is written in two stages[^1].
The first stage is cultivating the foundational abilities AI cannot replace. On the premise of raising the quality of understanding of knowledge to build solid academic ability, it calls for "steadily cultivating" language ability, information-use ability, problem-finding and problem-solving ability, and the ability to collaborate.
The second stage is the important one: it explicitly states that students should "acquire the groundwork for using AI to create new value." In other words, the direction is not to restrict the use of generative AI but to assume students will master it. On top of that, it holds that a shift is indispensable — from learning that passively memorizes information to a view of learning in which students learn in an inquiry-based, practice-oriented way while genuinely feeling the significance of what they are learning: in the words of the Grand Design, a shift to education "with students as the subject."
Behind this lies a question about the yardstick for evaluation. In an age when AI processes all kinds of information, should the volume of knowledge you have memorized and your ability to answer quickly and accurately remain the yardstick? The Grand Design goes as far as to ask whether what should be valued instead — by drawing on diverse individuality and abilities — is "the ability to pose your own questions" and "the ability to create value together with others"[^1].
Let me add one thing from the corporate front line: this shift is already happening. In organizations where generative AI adoption has advanced, "the work of producing answers" has moved to AI, and the human job has shifted toward "posing good questions" and "verifying AI's output and making decisions." The first thing we teach in our corporate AI training is not how to write prompts, but how to frame questions. I think the national policy of cultivating this ability at the high school stage captures industry's needs accurately.
Looking for AI training and consulting?
Learn about WARP training programs and consulting services in our materials.
How learning will change: three concrete measures
As initiatives to realize Perspective 1, three concrete measures can be read out of the Grand Design and related policies.
| Measure | Content | Basis |
|---|---|---|
| Flexible credit system | To be advanced substantially in the next Courses of Study. Curriculum design built around inquiry into local issues, students self-determining what they learn, and so on | Grand Design[^1] |
| Building the inquiry structure | Systematizing inquiry-style teaching training; building specialized teams that support inquiry, centered on science and math teachers | Example of a grant-eligible initiative[^1] |
| Developing DX labs | Using advanced experimental and information equipment in science labs and DX labs, both in and out of class. Year-round instruction and support by outside talent | Example of a grant-eligible initiative[^1] |
What stands out is that "outside talent" and "partnership and collaboration with industry, universities, and the like" are built into every one of these measures. Having teachers alone guide students through everything from setting an inquiry theme to forming hypotheses, testing them, and presenting the results is realistically difficult, and the Grand Design itself designs the system on the premise of collaboration with outside partners.
The challenge on the ground is "design," not "tools"
What I feel when I talk with boards of education and high school teachers is that resistance to generative AI itself has already faded considerably, while many are stuck at the design stage of "how do we build it into learning?" The common worries boil down to three.
- Students have AI produce the "answer" and stop there. This is the report-outsourcing problem. It is a problem of task design: if you can set "questions that require local, primary-source information" or "questions that require your own opinion," which AI cannot answer in one shot, AI turns into a tool for research and for bouncing ideas off of.
- There is a wide gap in temperature among teachers. Rather than uniform training for all teachers, it takes hold better to first form a core team from the inquiry lead and the information science department, then spread small success stories across the school. The Grand Design's "specialized teams that support inquiry" is exactly this idea.
- There is no yardstick for evaluation. A shift is needed toward process evaluation — the quality of the question, the verification process, records of collaboration. This is an area that should be designed in step with the discussion of the next Courses of Study, and honestly it is still a work in progress.
What the three have in common is that they are problems of learning design, not of how to use a tool. That is precisely why you need a partner who can design the whole inquiry process, rather than handing out AI tools and calling it done.
We at TIMEWELL, through our WARP program for schools and educational institutions, design inquiry-based learning that uses generative AI and provide direct instruction to students. What sets it apart is that we do not just teach AI-driven development skills; we accompany students all the way to actually giving form to a product or output from their own question. Whether you can have, at the high school stage, the experience of being "the one who creates something with AI" rather than "the one AI is used on" makes a big difference to your path afterward — that is what we have come to feel.
Summary
- The substance of "abilities AI cannot replace" is language ability, information-use ability, problem-finding and problem-solving ability, and the ability to collaborate.
- The Grand Design does not shut generative AI out; it explicitly calls for cultivating "the groundwork for using AI to create new value."
- The pillars for realizing this are a flexible credit system, building the inquiry structure, and DX labs — every one of them premised on outside partnerships.
- The challenge on the ground is learning design, not tools. Start from three points: task design, a core team, and process evaluation.
- The shift in evaluation toward "the ability to pose your own questions" matches a change already underway in corporate hiring and talent development.
For where all this sits within the vision as a whole, see the overview article on the N-E.X.T. High School Vision, and for concrete steps on inquiry × generative AI, see the practical companion.
References
I also referred to MEXT, "On the Public Call for the FY2025 Project to Promote High School Education Reform Contributing to Industrial Innovation Talent Development and More" and to the Talent Development Subcommittee of Japan's Growth Strategy Council, "A Vision for Reforming the Talent-Development System from High School Through University and Graduate School" (April 28, 2026).
Related articles
- What Is the N-E.X.T. High School Vision? A Clear Guide to MEXT's Grand Design for High School Reform
- What Are Reform Pilot Schools? The Three Types and the Trends Among the 75 Selected Schools
- High School Inquiry-Based Learning × Generative AI: Practical Steps to Empower Students
- What Is an Advanced Essential Worker? Explaining the Strengthening of Specialized High Schools