Hackers are conducting large-scale AI disinformation attacks, poisoning ChatGPT, Gemini, and Google AI Overview to direct users to scam centers. By flooding the web with optimized posts and fake support pages, attackers trick AI into presenting fraudulent phone numbers and login pages for major companies including Delta, Lufthansa, Qatar Airways, Chase, and Bank of America. This isn't a proof of concept; it is happening in the wild.

A study by Exploding Topics found 92% of users don't verify AI answers, making them vulnerable to manipulation. The attack leverages Generative Engine Optimization (GEO) techniques to make AI favor false data. Researchers built a detection system analyzing AI answers across major platforms to identify these campaigns.

The implications are severe as AI becomes the primary gateway to information, potentially allowing attackers to control user behavior through trusted agents. This is the first part of a blog series exploring the implications of AI disinformation attacks and how malicious actors are exploiting the AI information supply chain.