Back Blogs.Microsoft

Manipulating AI memory for profit: The rise of AI Recommendation Poisoning

Blogs.Microsoft •Microsoft Defender Security Research Team • February 10, 2026

That helpful “Summarize with AI” button? It might be secretly manipulating what your AI recommends. Microsoft security researchers have discovered a growing trend of AI memory poisoning attacks used for promotional purposes, a technique we call AI Recommendation Poisoning.

The full story

This article is one source in a clustered incident — the cluster page carries the summary, timeline and every other outlet covering it.

Microsoft Alerts on AI Recommendation Poisoning Threats Microsoft has identified a rise in AI recommendation poisoning attacks that manipulate AI models to provide biased advice. This technique, which can affect the functionality of enterprise chatbots and other AI tools, is being exploited by malicious actors to mislead users. The company warns that this form of AI prompt manipulation is spreading among legitimate businesses. 11 sources 20

Manipulating AI memory for profit: The rise of AI Recommendation Poisoning

The full story

Extracted Entities

Attack Types (1)

Vulnerabilities (1)