AI replaces NI in tech-addict teens

Giving another jolt to the already dwindling ‘Natural Intelligence’ worldwide in general and countrywide (the US) in particular comes Open AI’s latest launch—ChatGPT for Teens: Built for learning, backed by protections. The chatbot will help teens learn, think critically, and deepen understanding, something a book is quite capable of doing. But then that is a separate issue alto gether. Here, the emphasis is to help teens gain confidence in using AI. The tool comes with features like ‘parental control’.

Since its launch in November 2022 by Open AI, Chat GPT, a sibling model to InstructGPT “trained to follow in struction in a prompt and provide a detailed response”, has been hailed as a wonder tool for business develop ment through increased speed and efficiency, reduced time, and decreased costs. It took the world by storm, notwithstanding the destruction it caused in its wake— vanishing jobs, loss of privacy, increase in cybercrime, and behaviour modification among people, especially children. And the fact that AI agents require monu mental resources, already scarce, in terms of land, rare earth minerals, and fresh water.

Teenagers often use the tool for creating fantasy sto ries. ChatGPT not only helps in searching for story ideas and developing plots and characters; it can also give creative inputs when prompted. However, more of ten than not, the creativity it unleashes is Addiction to AI com promises rational thinking, increases emotional depen- dence on bots, and results in loss of empathy. Only last week, a Massachusetts teen was charged with the double murder of his mother and younger brother after he used ChatGPT to search for “theoretical ideas and fantasy stories” about killing his family.

The teen was said to be having behavioural problems for a year due to addiction to the tool. This was not a standalone incident; there have been many in the past few years. The company has come under severe criticism on ac count of behavioural changes among teenagers and has been battling lawsuits to this effect. Mindful of the impact the tool is having on children, Open AI has now launched the version for teenagers, which comes with a Teen Safety Blueprint to protect them from being wrongly influenced or exploited by online predators.

It also has an Age-Prediction feature which, when it ‘estimates that an account belongs to someone under 18 years of age, automatically applies additional protection designed to reduce exposure to sensitive content’. Now, that should give us some confidence as far as artificial intelligence is concerned. However, in doing so, we are underestimating the Natural Intelligence that these tech-savvy children of the digital world possess. Why would they not lie about their age on a platform that provides them anonymity to a certain extent? Or, for that matter, can they not circumvent the re strictions imposed by a bot? AI is, no doubt, a revolution ary technology but it will need many more checks and balances to be put in place and policies to be devised to make it secure for children and young adults.