--- title: "The Cost of Letting AI Think for Our Children" date: 2026-10-14 author: "Robert A. Lee" featured_image: "https://sqmagazine.co.uk/wp-content/uploads/2026/10/vasyl-zahorodniuk-teaching-children-think-themselves-age-ai-v3.jpg" categories: - name: "Artificial Intelligence" url: "/artificial-intelligence.md" tags: - name: "SP" url: "/tag/sp.md" --- # The Cost of Letting AI Think for Our Children Education has always required an investment whose returns arrive slowly. Children spend years learning to read closely, express themselves and work through problems before those abilities acquire a professional value. Artificial intelligence now offers the appearance of the same result in seconds. It can supply the explanation, complete the calculation and write the essay. The temptation is to ask whether the years of preparation are still necessary. That question confuses access to knowledge with the ability to use it. An education gives people the means to judge an answer, recognize a faulty argument and decide which questions deserve their attention. As information becomes cheaper and more abundant, those distinctions become more important. While business and political leaders debate whether AI represents an existential threat or the next great advance in productivity, Ukrainian-American entrepreneur Vasyl Zahorodniuk is concerned with a more immediate problem: how to educate [children who will grow up with it](https://sqmagazine.co.uk/generation-alpha-statistics/). Through an initiative involving Zahorodniuk Foundation and Quanta Tech Systems LLC, he is developing an educational approach intended to preserve independent thought while preparing pupils to use increasingly capable technology. The distinction between an adult user and a young child is central to his argument. Adults in their twenties, thirties and forties generally bring years of schooling and practical experience to an exchange with AI. They have at least some basis for questioning an answer. A professional can delegate a familiar task while retaining the knowledge required to assess the result. Children between two and ten are still acquiring that foundation. Their curiosity often runs ahead of their judgment, as it should. Education gives that curiosity direction through conversation, experience and sustained effort. Supplying an answer too readily can remove an opportunity to learn how to arrive at one. Language deserves particular attention. It is both a means of communication and a foundation for learning. The responsive exchanges between a young child and a caring adult help shape brain architecture, supporting early language and social skills on which more complex cognitive abilities depend. A child’s effort to describe an experience or explain an idea is therefore part of the work of development – not an inconvenience to be eliminated by a more efficient system. The concern is what happens when assistance consistently takes the place of that effort. A polished response can conceal an incomplete understanding. A child who becomes adept at obtaining explanations may still struggle to construct one. If schools assess only the finished product, they risk rewarding the capability of the software while overlooking the progress of the pupil. ![Primary school pupils using tablets in a classroom with their teacher](https://sqmagazine.co.uk/wp-content/uploads/2026/10/children-ai-classroom-tablets.jpg) [Restricting AI in the classroom](https://sqmagazine.co.uk/ai-in-education-statistics/) can preserve time for independent work. It cannot, by itself, prepare children for the technology they encounter elsewhere. Phones, family computers and services used outside school make this a responsibility shared by educators and parents. Children need boundaries, but they also need an understanding of why those boundaries exist. We should not hide our children from AI. We should educate them. Zahorodniuk is particularly opposed to what he describes as a Tamagotchi-style approach: AI products built around a child’s attachment to a digital companion. His objection is to making the relationship with the product an end in itself. In his proposed educational model, technology would have a defined purpose and a limited place, leaving human conversation, play and independent discovery at the center of childhood learning. Recent research gives the debate a more concrete basis, although it does not settle the question of childhood development. In Your Brain on ChatGPT, a preliminary study led by Nataliya Kosmyna, researchers used electroencephalography to examine participants during essay-writing tasks. Those using an AI assistant showed weaker neural connectivity than those writing without tools. In a later session, participants who switched from AI-assisted to unaided writing showed reduced connectivity in certain frequency bands, which the researchers interpreted as under-engagement. The findings should not be mistaken for proof of permanent brain damage. Nor does an adult writing study establish that AI causes developmental impairment in children. It does, however, raise a question educators cannot dismiss: what happens to learning when a tool performs much of the mental work a task was intended to require? For an adult, that may mean delegating an ability already acquired. For a child, it may mean losing opportunities to acquire the ability in the first place. The developmental concern is not that every encounter with AI is harmful. It is that habitual substitution could deprive children of practice they still need. Zahorodniuk’s response is a proposal for a new kind of education, rather than a choice between unrestricted access and prohibition. The program under development would place language, reasoning and independent work before technological convenience. Children would be expected to attempt an explanation, consider evidence and express a conclusion in their own words. AI could then help them examine or improve their work without doing the essential thinking on their behalf. For younger children, this would mean adult-led, age-appropriate activities rather than unrestricted conversations with a chatbot. Teachers and parents would remain responsible for deciding when a tool serves a learning objective and when it removes a useful challenge. The intention is to make the program compatible with established educational frameworks, including schools offering International Baccalaureate programs. That compatibility is a development objective, not a claim of IB endorsement. Its value would ultimately have to be demonstrated in pupils’ progress: what they understand, what they can explain and what they can do when assistance is withdrawn. ## Who Is Vasyl Zahorodniuk? Vasyl Zahorodniuk is a Ukrainian-American entrepreneur, investor and security specialist whose background spans Ukraine, the United States and Spain. Through Quanta Tech Systems LLC, he works in cybersecurity and technology development. His educational initiative extends that interest to the habits and abilities people need to use technology responsibly. Alongside his business activities, Zahorodniuk pursues research in mathematics and the social sciences. These interests inform an approach to education concerned with both the discipline of reasoning and the human circumstances in which learning takes place. His commitment to educational philanthropy in the United States and internationally includes his involvement in Rotary and the work being developed through Zahorodniuk Foundation. The underlying conviction is that children’s intellectual development deserves sustained investment, regardless of how capable the machines around them become. The school of tomorrow will have extraordinary tools at its disposal. Its success should still be judged by what its pupils can think through for themselves.