ChatGPT-maker OpenAI launches Teen Safety Blueprint for India, focuses on 'safety' over 'privacy'

At a time when AI companies are under pressure to offer safe chatbot conversation experiences to kids, OpenAI, the company behind ChatGPT, has released safety blueprint for teens in India. The company emphasises that it aims to keep safety over privacy for younger users. “We're taking steps to strengthen protections for Indian teens through age prediction, age-appropriate policies, and parental controls, informed by conversations with policymakers, regulators, educators, and child safety experts, as well as mental health and digital well-being practitioners,” the company says in the report.According to OpenAI, the new protections come in addition to features already available for all users. These include in-app reminders during long sessions to encourage breaks; safeguards that direct users to real-world resources if the company’s systems detect they have expressed suicidal intent. The company also offers “ways to escalate risks of physical harm to others for human reviewers who can take action; and industry-leading prevention of Al-generated child sexual abuse material (SAM) and child sexual exploitation material (CSEM). In particular, we constantly improve our CSAM/CSEM detection methods and report confirmed CSAM to relevant authorities.” OpenAI claims that it aims to ensure that all teens using Al receive age-appropriate protections by default, and that parents and educators can deploy additional protections to personalize how teens use Al. OpenAI’s blueprint acknowledges a hurdle unique to the Indian market: the prevalence of shared technology. According to the RATI Foundation’s Ideal Internet Report 2024-25, approximately 62% of Indian teens access the internet via shared family devices.The core of the new framework is the concept of “age-aware AI behaviour.” OpenAI argues that AI responses should not be “one-size-fits-all,” advocating for a system where ChatGPT recognises the developmental stage of the user.“We believe that Al companies should identify teens on their platforms using privacy-protective, risk-based age estimation tools to distinguish between teens and adults. These tools should minimize the collection of sensitive personal data, while still effectively distinguishing users under the age of 18 (U18). Where possible, these methods might also rely on operating systems or app stores to determine a user's age,” the company says.Furthermore, the framework introduces enhanced Parental Controls, allowing parents to link accounts, set “blackout hours” for AI usage, and remotely disable chat history and memory features.