OpenAI Japan: Securing AI for Teens

OpenAI Japan Commits to Teen Safety in the Age of AI

OpenAI Japan recently unveiled the Teen Safety Plan, an ambitious initiative aimed at ensuring that young users can interact with generative AI safely. In a country where the adoption of these technologies by adolescents is rapidly expanding, particularly for educational, creative, and everyday activities, this initiative is of paramount importance. As this generation grows in symbiosis with AI, it becomes essential to ensure that these tools are developed with a particular focus on their safety and well-being.

Generative AI already plays a significant role in various aspects of life, from learning to artistic expression, as well as in everyday tasks that facilitate academic, professional, and personal success. On a larger scale, it has the potential to accelerate scientific advancements and provide solutions to complex societal problems.

However, like any powerful technology, AI carries risks, especially for young users. These include exposure to misinformation, inappropriate content, and psychological stress. OpenAI adopts an approach where the safety of adolescents takes precedence, even if it means compromising on convenience, privacy, or freedom of use.

Pillars of the Teen Safety Plan

• Age-Appropriate Advanced Protections
  OpenAI will implement an age estimation system based on privacy and risk to differentiate adolescents from adults and provide specific protections for each group. Appeal mechanisms will be available for users to contest age assessments deemed incorrect.

• Enhanced Safety Policies for Under-18s
  OpenAI will strengthen its measures to ensure that AI does not promote or glorify self-harm, suicide, or generate explicit sexual or violent content. AI should also not encourage dangerous behaviors or reinforce harmful body images. Responses will be tailored to the developmental stage of young users, and AI must not assist minors in hiding risky behaviors or health issues from their parents or guardians.

• Expanded Parental Controls
  Tools such as account linking, privacy settings, usage time management, and tailored alerts will allow families to customize protections according to their specific needs.

• Well-Being-Centered Design
  In collaboration with clinicians, researchers, educators, and child safety experts, OpenAI will continue to enhance features such as pause reminders and pathways to real support, while also researching the impact of AI on adolescent mental health and development.

• Existing Protections
  These efforts will strengthen the protections already in place in ChatGPT, including:

  • Built-in reminders to encourage breaks during prolonged use
  • Safety systems that detect signs of self-harm and direct users to real resources
  • Multi-layered safety systems and abuse monitoring
  • Advanced prevention of AI-generated child sexual exploitation material

Collaboration with Japanese Society

OpenAI supports the Japanese approach that aims to balance robust protections for minors with responsible access to technology. The company will continue to engage transparently with parents, educators, researchers, policymakers, and local communities to enhance its efforts. Protecting adolescents in the age of AI is a shared responsibility, and such protections should become a standard in the industry.

"Protecting the safety of a generation growing up with AI is a shared responsibility of society, including businesses, governments, educators, and families. OpenAI is committed to working closely with stakeholders in Japan to create an environment where young users can learn, create, and develop their potential with confidence." — Kazuya Okubo, Head of Policy and Partnerships, OpenAI Japan

OpenAI will continue to advance these efforts as part of its broader commitment to safe and responsible AI.