OpenAI has launched a separate ChatGPT mode for teens. How will it work?

This is reported by The Verge.

The mode is activated automatically if the user has indicated an age of under 18 or the system itself has decided that the account is probably being used by a minor. To assess the age, ChatGPT takes into account not only the age indicated by the user himself, but also how long the account has existed, what time of day the person usually uses the service and their usage patterns over time. If an adult is mistakenly “recorded as a teenager” by ChatGPT, they will be able to confirm their age through the Persona service. Depending on the country, they may be asked for a selfie or a document.

For teens, ChatGPT will place stricter restrictions on graphic violence, dangerous viral challenges, and content that promotes extreme beauty standards, unhealthy diets, or body shaming. OpenAI’s separate rules for users under 18 also include stronger safeguards for conversations about self-harm and other risky behavior. Romantic and sexual role-playing games are also subject to additional restrictions.

Teenagers will also be reminded more often that they are communicating with AI, will be offered to take a break after prolonged use, and before some photo uploads, ChatGPT may ask if there is any sensitive information in the picture that is better not to share.

Separately, OpenAI has taken on homework. The new “responsible homework reminders” are supposed to recognize situations where a teenager is simply trying to get a ready-made answer and suggest switching to Study Mode. There, ChatGPT guides the user through the task step by step: asks leading questions, explains the material, and can check whether the person has understood it.

There is also a study hours function. A teenager or parents can set certain hours when all new chats will automatically open in study mode. And so that a teenager cannot use ChatGPT at all at a certain time, there are quiet hours.

Despite the name “parental controls,” an adult can’t read a teen’s chat history or monitor their activity in real time. In certain serious cases, OpenAI can send parents a notification, such as if the system and moderators have detected a risk of self-harm. However, it won’t contain details of the conversation itself.

Previously, dev.ua wrote that Meta is also strengthening protection for minors when communicating with AI. The company has started notifying parents if Meta AI detects signs of self-harm or suicidal thoughts in a teenager's chats, and parental content restrictions on Instagram now extend to conversations with the chatbot.

Google and Character.AI reach settlements in teen suicide cases after conversations with their chatbots

On the topic

Google and Character.AI reach settlements in teen suicide cases after conversations with their chatbots

ChatGPT will force teens to take breaks from prolonged use of AI

On the topic

ChatGPT will force teens to take breaks from prolonged use of AI

Read the country's main IT news in our Telegram

On the topic

Read the country's main IT news in our Telegram