Updating our Model Spec with teen protections
OpenAI has announced an update to its Model Spec by introducing new "Under-18 Principles" to govern model behavior for younger users. Released on December 18, 2025, these guidelines aim to ensure ChatGPT provides age-appropriate assistance informed by developmental science. The changes strengthen safety guardrails and offer clearer expectations for model responses during higher-risk scenarios.
Key Takeaways
- On December 18, 2025, OpenAI announced updates to its Model Spec to incorporate new "Under-18 Principles" aimed at protecting younger users.
These guidelines outline how ChatGPT must deliver safe, age-appropriate support designed around developmental science.
- By establishing these explicit rules, the company seeks to align AI behavior with the specific developmental needs of teenagers using the platform.
The revised spec strengthens existing guardrails and provides explicit instructions for model conduct in higher-risk situations.
- This initiative expands on OpenAI's ongoing efforts to enhance safety measures for adolescent interactions across ChatGPT.
This update demonstrates how AI safety specifications can be tailored directly to specific user demographics and risk profiles.
- OpenAI updated its Model Spec to establish new "Under-18 Principles" for ChatGPT users.
The revised principles provide safe and age-appropriate guidance informed by developmental science.
- The update enhances guardrails and specifies expected model actions in higher-risk situations.
Stats & Key Facts
- #OpenAI has announced an update to its Model Spec by introducing new "Under-18 Principles" to govern model behavior for younger users.
- #On December 18, 2025, OpenAI announced updates to its Model Spec to incorporate new "Under-18 Principles" aimed at protecting younger users.
- #OpenAI updated its Model Spec to establish new "Under-18 Principles" for ChatGPT users.
On December 18, 2025, OpenAI announced updates to its Model Spec to incorporate new "Under-18 Principles" aimed at protecting younger users. These guidelines outline how ChatGPT must deliver safe, age-appropriate support designed around developmental science. By establishing these explicit rules, the company seeks to align AI behavior with the specific developmental needs of teenagers using the platform.
The revised spec strengthens existing guardrails and provides explicit instructions for model conduct in higher-risk situations. This initiative expands on OpenAI's ongoing efforts to enhance safety measures for adolescent interactions across ChatGPT. This update demonstrates how AI safety specifications can be tailored directly to specific user demographics and risk profiles.
OpenAI updated its Model Spec to establish new "Under-18 Principles" for ChatGPT users. The revised principles provide safe and age-appropriate guidance informed by developmental science. The update enhances guardrails and specifies expected model actions in higher-risk situations.
For more details please read the original article at OpenAI.
Continue Learning
Comments
Comments appear only after moderation. Your email identifies your submission to the moderator and is never displayed here.
No approved comments yet.