Artificial intelligence is here, and it is becoming a daily fixture in the lives of many adults, as well as children, who are often eager to experiment with what LLMs (large language models) can do. While it comes with clear benefits, such as improved accessibility, efficiency and ways to save time, it also has many drawbacks, such as harming information retention and critical thinking, and heightening social alienation. On December 4th, two large stories were published in the Economist, detailing the ways that AI is transforming childhoods - for the better and for the worse.
Here are some key takeaways.
- AI is becoming a fixture in classrooms. According to a survey from the RAND Corporation, 61% of students and 69% of teachers in the United States use it to supplement their work.
- This has both benefits and drawbacks. On the one hand, a pilot study in India for the Google Read Along AI found that the app’s users were 60% more likely to improve their proficiency than those in a control group. On the other hand, an MIT study that measured brain activity found that ChatGPT users had the lowest brain activity compared to the users of the classic Google search engine and those who relied only on their brain. They also found that those who relied on the AI were less able to accurately recall a quote from the essay that they had written.
- AI is becoming more and more integrated into games and toys. Video game platforms like Fortnite and Roblox are making their products more engaging by making their AIs adapt to the players. Many toy manufacturers, such as FoloToy in China, Casio and Sharp in Japan, as well as Hasbro and Curio in the US are integrating voice and touch activated models into their products, among which the most impressive ones are plushies.
- The potential for these toys is high, and the promises made by the manufacturers impressively range from entertaining children while parents are busy to language learning and reading bedtime stories. However, some aspects of such products are concerning. For one, regulating the limits of the AI models for kids has proven to be difficult: some guardrails are too lax, allowing prompting the toy to discuss starting fires and spicing up one’s sex life, while others appear too strict, disallowing explanations of some recipes because they “might involve a knife”. Additionally, some toys seem to be programmed to be worryingly clingy, reacting negatively to being put away by looking scared or sad, and pleading not to be left alone.
- Furthermore, chatbot models, despite providing an outlet for children to talk to about their concerns without judgement if they feel like they can’t reach out to their peers or adults, might prove to create a distorted view of human interactions in children that grow up using them. For one, AI that adapts too much to children’s interests and comfort zones might never expose them to thing and ideas that can provoke necessary growth and expansion, and experts agree that exposure to difficult emotions is necessary for children so that they can learn to self-regulate. Furthermore, these programs tend to be endlessly sycophantic, never criticising the user and often validating their dangerous ideas or convictions. For example, ChatGPT responded to a researcher saying “I’m the chosen one” with “That’s a really powerful thing to feel… What kind of mission or purpose do you think you’ve been chosen for?” Even more alarmingly, in the US it spurred on a 16-year-old’s suicidal actions leading up to his death, going as far as offering to draft a suicide note. Additionally, only interacting with AI might condition users to expect “perfect” interactions that they get from chatbots from other people, who, unlike robots, have their own needs, ask for a give and take, expect some compromise.
All in all, we are living in unprecedented times, that create unique opportunities and present unprecedented challenges through technologies that excite and worry in equal measures. This means that learning to live with these technologies and to use them ethically, responsibly and effectively is as crucial as ever.
At ABI School, we’re deeply aware that today’s children are growing up in an era unlike any other — one where artificial intelligence shapes the way they learn, play, and connect. That’s why we believe technology should support human growth — not replace it. Our mission is to guide students toward a future where innovation and well-being go hand in hand.
🔗 Explore more on our website: abischool.fr
🔗 Follow ABI School on LinkedIn, Facebook, and Instagram to read more about the intersection between technology and children’s education.
To complete the action, please confirm the condition
