High School Textbook Screening: Be Careful Not to Use AI Incorrectly

Editorial

15:29 JST, March 26, 2026

Artificial intelligence is a convenient tool, but if used incorrectly in learning environments, it could hinder the acquisition of thinking ability and expressive power. It is therefore essential to first thoroughly understand its characteristics and risks.

The Education, Culture, Sports, Science and Technology Ministry has released the results of the screening of textbooks to be used primarily by second-year high school students from the 2027 school year. This was the second screening based on the current curriculum guidelines. Applications were made for 200 items from 22 publishers across eight subjects, not including specialized subjects, and 196 items from 21 publishers were approved.

Of these, 67 items across all eight subjects included content addressing the characteristics and risks of AI. Descriptions related to information literacy, such as elections and disinformation on social media, were also prominent.

How should modern society deal with rapidly developing AI and social media, with related issues such as defamation becoming increasingly apparent? Many textbook publishers are likely showing signs of a growing awareness of this question.

Chikumashobo Ltd.’s Japanese language textbook features a section titled “How to deal with generative AI,” asserting that “without the reflections and expressions of humans themselves, the power to appeal to readers’ intellect and sensibilities would not be able to exist in texts.”

The percentage of high school students using generative AI has already reached 46%. In educational settings, there is a reality that high school and university students are having generative AI write their assignments and papers instead of writing them themselves, leaving teachers struggling to deal with the issue.

The value of texts lies in the fact that the writer’s personality is built through their own thinking and careful choice of expression. It is quite natural that writing produced by AI — which has no experience, such as reading, traveling and romance — is hollow.

It is important not to forget that AI is, after all, merely a tool. It is essential for teachers to enhance their knowledge of AI and apply it to their instruction, too.

Daiichi Gakushusha Corp.’s English textbook noted the dangers of disinformation, citing how AI-generated images caused confusion during disasters. A textbook also addressed “hallucinations” — instances in which AI provides plausible answers that contradict the facts.

Using AI in a manner that apparently replaces human thought could hinder the development of “intellect.” Recently, many young people reportedly refer to the conversational AI service “ChatGPT” as “Chappy,” and use it to discuss their worries, or as a conversation partner.

Worrying is also an important process of personality development. AI should not be allowed to make decisions about a person’s own judgments and actions.

A symposium was held at Keio University to consider ethics in the digital age. It is hoped that discussion will be deepened across generations about how to use AI and social media properly.

(From The Yomiuri Shimbun, March 26, 2026)