AI Result Poisoning

T1608.007

Sub-technique of T1608 Stage Capabilities.View on attack.mitre.org

About this technique

Adversaries may manipulate publicly accessible web content to influence generative AI systems or AI-powered search engines into retrieving, citing, or recommending attacker-controlled information.

Unlike traditional SEO Poisoning, which primarily attempts to influence the ordering of conventional search results, AI Result Poisoning targets systems that synthesize information from multiple sources and provide a direct answer to the user. Adversaries may create malicious or misleading websites, publish fabricated claims, generate supporting articles, or establish multiple apparently independent sources that reinforce the same information. These sources may then be retrieved by AI systems using web search or retrieval-augmented generation and incorporated into generated responses.

Adversaries may use this technique to recommend or link to phishing sites, fraudulent online stores, malicious downloads, impersonation pages, or other attacker-controlled infrastructure. It can also be used to promote misinformation or fabricated claims without requiring access to the underlying AI model, system prompt, or application infrastructure. Because the malicious content can resemble ordinary marketing, editorial, or informational material, traditional prompt injection and malware-focused detections may not identify the activity.

Detection rules0

Rules on DetectionCode tagged with T1608.007.

Sigma0

No Sigma rules are mapped to this technique yet.

Splunk0

No Splunk rules are mapped to this technique yet.

Groups0

None recorded.

Software0

None recorded.

Campaigns0

None recorded.

Procedure examples0

No procedure examples are recorded for this technique.

References6

  1. Digiday AI Result Poisoning Open source
    Sam Bradley. (2026, July 6). WTF is AI poisoning?. Retrieved September 18, 2026.
  2. Lasso Security AI Engine Poisoning Open source
    Eliya Saban. (2026, June 24). Exploiting GEO to Push Harmful Claims into AI-Generated Answers. Retrieved September 18, 2026.
  3. Netcraft AI Summaries Open source
    Ginny Spicer. (2026, August 3). Threat Actors are Finding Their Way into Your AI Summaries. Retrieved September 18, 2026.
  4. darkreading ai result poisoning Open source
    Alexander Culafi. (2025, December 10). ClickFix Style Attack Uses Grok, ChatGPT for Malware Delivery. Retrieved September 18, 2026.
  5. darkreading geo poisoning Open source
    Alexander Culafi. (2025, December 10). ClickFix Style Attack Uses Grok, ChatGPT for Malware Delivery. Retrieved September 18, 2026.
  6. dev.to Abusing AI Chatbot Recommendations Open source
    Dennis Kim. (2026, May 27). Cryptojacking Abusing AI Chatbot Recommendations — A New Delivery Vector Beyond Search Poisoning. Retrieved September 18, 2026.

Data from MITRE ATT&CK® (Enterprise). ATT&CK® is a registered trademark of The MITRE Corporation.