---
type: "Evidence Item"
title: "Keeping your data safe when an AI agent clicks a link"
description: "Learn how OpenAI protects user data when AI agents open links, preventing URL-based data exfiltration and prompt injection with built-in safeguards."
resource: "https://openai.com/index/ai-agent-link-safety"
tags: ["appendix-iii", "vendor", "openai"]
timestamp: "2026-01-28"
category: "vendor"
publisher: "OpenAI"
cope_score: 46
confidence: 0.9
---

# Keeping your data safe when an AI agent clicks a link

# Claim

Learn how OpenAI protects user data when AI agents open links, preventing URL-based data exfiltration and prompt injection with built-in safeguards.

# Relevance

Appendix III, section two: vendor threshold and platform capability evidence

# Oracle Verdict

This is a low-signal vendor radar item. Keep it as context only unless a later benchmark, deployment, procurement change, or labour-market datapoint turns it into direct Appendix III evidence.

# Metadata

* Publisher: OpenAI
* Category: vendor
* Sector: Cybersecurity
* Capability: Agent platform and API infrastructure
* Cope score: 46
* Confidence: 0.9

# Related Concepts

* [Live evidence index](index.md)
* [Thesis](../thesis.md)

# Citations

[1] [Keeping your data safe when an AI agent clicks a link](https://openai.com/index/ai-agent-link-safety)