Our Approach to Protecting Children's Safety
Original title:Our Approach to Child Safety
AI summary
Runway follows Thorn's 'Safety by Design for Generative AI' principles to protect children from AI abuse throughout the process. Starting from model development, it uses hash matching, child safety classifiers, and LLM review to ensure training data does not contain sexual content involving minors, and conducts red-team testing to identify vulnerabilities. After product deployment, sexual content involving children is explicitly prohibited, a multi-layer detection system is used to scan user content, all flagged content is manually reviewed and reported to the U.S. National Center for Missing & Exploited Children (516 submissions in 2025). It also implements C2PA provenance signals to track content generation and continues to work with industry organizations to address threats.
Key Takeaways
- Runway follows Thorn's 'Safety by Design for Generative AI' principles to protect children from AI…
- Starting from model development,it uses hash matching,child safety classifiers,and LLM review to…
- After product deployment,sexual content involving children is explicitly prohibited,a multi-layer…
- National Center for Missing & Exploited Children (516 submissions in 2025).
- It also implements C2PA provenance signals to track content generation and continues to work with…
Quick Facts
- Published
- Updated
- Source
- Runway:News(网页)
- Category
- Industry
- AI score
- 65
这不是模型发布,但 Runway 首次详细公开了从训练数据到举报 NCMEC 的完整儿童安全链路,有 516 份举报记录,对关心 AI 合规和安全设计的产品人是个重要参考。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: Our Approach to Protecting Children's Safety. May 8, 2026. https://new.nat6.net/en/news/cmoxivzb2023rsllhka40ewoy/