AI giant OpenAI says it is launching a new mode of ChatGPT specifically for teens with stronger restrictions and an emphasis on safety as concerns over minors’ use of the chatbot grows.
The safer mode for under 18s will turn on by default when ChatGPT is unsure of a user’s age – or the user themself says they are between 13 and 17.
There will be more restrictions on content and parents will have greater controls.
OpenAI say they want to ” treat teens like teens”, but also make sure that “they’re not exposed to material that they shouldn’t be exposed to”.
Greater guardrails
Parents will also have greater control, being able to set ‘Quiet Hours’ and receiving notifications when sensitive topics are discussed.
OpenAI gives eating disorders as an example of a topic it would flag.
ChatGPT for Teens will also change the way it helps with homework, focusing on teaching and learning rather than spitting out the correct answer, according to OpenAI.
A recent study conducted jointly by Stockholm University and the University of Hong Kong shows that, while children who use AI tend to score higher on homework, their exam scores take a hit compared to those who do not use AI.
‘Study Mode’ will attempt to give hints and guide students to work out the answer for themselves.
“AI should expand educational opportunity,” say OpenAI, and teen mode will help “support moments outside the classroom”.
YOU MAY ALSO LIKE: Meta in the spotlight as landmark child-safety trial begins
Man or machine
The teen mode will also have more guardrails to prevent it from sounding like a real person, say OpenAI.
In a statement on OpenAI’s website, ChatGPT’s new teen mode “goes beyond blocking romantic or sexualized roleplay” and will “not use romantic language, encourage emotional dependence, or imply that it has feelings or consciousness”.
OpenAI says it wants this ChatGPT to “feel personal without blurring the line between a useful tool and a human relationship”.
Do teens and tech mix?
Scrutiny against AI chatbots has been rising after stories of teens self-harming and even committing suicide, allegedly due to taking to chatbots.
Google’s AI chatbot Gemini, as well as OpenAI’s own ChatGPT, have both been taken to court in separate incidents involving minors taking their own lives.
The parents of Adam Raine, a US teen who took his own life, say that ChatGPT discouraged him from talking to them about his problems and even provided information on suicide methods.
This case, as well as several others, are still ongoing.
There have also been several reported murders whose perpetrators showed excessive AI use.
MORE AI NEWS: UN chief calls for swift global regulation of AI, warns of growing risks
The future of youth AI use
Although many countries are introducing restrictions or bans on minors using social media, the same cannot be said for AI.
Meta, a tech giant also in the AI race, is currently facing a major US child-safety lawsuit.
Meta is accused of deliberately designing its social media apps Facebook and Instagram to keep young people hooked.
The outcome of the trial may well affect how young people’s use of AI is also viewed.
READ NEXT: South Korea condemns North Korea’s firing of ballistic missiles
Ian Trinidad 
Gabrielle Wilkinson 
Carmela Calling