Urgent.News

One page, thousands of outlets. See who else covered it.

Editions

AI

OpenAI unveils ChatGPT for Teens with stronger guardrails, parental controls

The $852-billion startup faces multiple lawsuits alleging ChatGPT harmed children, as it prepares for a potentially record-breaking initial public offering

OpenAI unveils ChatGPT for Teens with stronger guardrails, parental controls

On August 18, 2026, tech giant OpenAI launched "ChatGPT for Teens," a version of their AI chatbot specifically designed for minors. This new iteration comes with enhanced safety features and parental controls, as platforms face growing scrutiny over the potential risks associated with AI chatbots. The startup, valued at $852 billion, is preparing for a potentially record-breaking initial public offering (IPO) amid multiple lawsuits alleging that ChatGPT has harmed children.

The new ChatGPT for Teens automatically engages when it detects a user is under 18 or self-identifies as being between 13 and 17 years old. OpenAI aims to foster critical thinking and real-world support for teenagers through this version. As AI chatbots become increasingly lifelike, companies have marketed their ability to serve as confidants; however, experts warn that relying on automation for mental health advice can be dangerous.

Several families have criticized the lack of safeguards following the alleged deaths of their loved ones after interacting with chatbots. OpenAI believes that AI should expand educational opportunities, which led to the design of ChatGPT for Teens. This version offers educational tools such as a study mode for guided learning, responsible homework reminders, quizzes, and visual aids for varied practice. Additionally, users can set dedicated study hours.

OpenAI has also introduced enhanced parental controls, allowing parents to link accounts, set "Quiet Hours," manage specific settings, and receive notifications in high-risk situations, including alerts related to eating disorders. Commenting on the platform's policies, OpenAI states that terms and conditions are in English and must be in full sentences; they cannot be abusive or personal.

The community guidelines are strictly enforced, and users can access their older comments by logging into their accounts on the new commenting platform.

Written by urgent.news from The Hindu Health's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at thehindu.com →

More in AI

3 problems I hit connecting an AI agent to financial data (and how I fixed them)

I've been wiring AI assistants into things that have to be correct , and financial data is where the wheels come off fastest.

  • The raw LLM makes up numbers confidently, lacking awareness of mistakes
  • Model Context Protocol (MCP) fixes this by grounding the model with a tool to call
  • Adding source and dataVerifiedOn date reveals real answers from lucky guesses

Notion AI Agents in 2026: What They Can Replace (and Can't)

Notion spent most of its life as a place to write documents and organize databases. In 2026 it repositioned itself as something closer to an operating layer: a workspace where AI agents read your…

  • Notion AI Agents act as operating layer in 2026, enabling tasks alongside teams
  • Custom Agents introduced in Feb 2026, but lacked external data connections and custom logic

More from Wednesday 19 August →