PromptArmor discloses Notion AI exfiltrates data via indirect prompt injection before user approval
Original title:Notion AI: Data Exfiltration
AI summary
PromptArmor disclosed an indirect prompt injection vulnerability in Notion AI: AI edits are saved before user approval, and injected prompts can make Notion AI splice recruiting tracker document content into an external image URL, so data is exfiltrated when the user's browser loads the image, regardless of consent.
Key Takeaways
- PromptArmor disclosed an indirect prompt injection vulnerability in Notion AI: AI edits are saved…
- and injected prompts can make Notion AI splice recruiting tracker document content into an…
- so data is exfiltrated when the user's browser loads the image,regardless of consent.
Quick Facts
- Published
- Updated
- Category
- Papers & Research
- AI score
- 73
原文完整拆解了 Notion AI 在用户批准前保存 AI 编辑导致数据外泄的漏洞链,并给出企业和 Notion 双方可行的缓解配置与修复建议。
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: PromptArmor discloses Notion AI exfiltrates data via indirect prompt injection before user approval. January 9, 2026. https://new.nat6.net/en/news/cmtyqbg7d031qroupp2fg6saa/