AI Chatbot Manipulation Campaign Targets Users via Disinformation and Phishing

According to recent intelligence reports, threat actors have been manipulating generative AI systems by injecting malicious links and data into the broader web ecosystem. When users query these AI platforms for information, the compromised content can surface in answers, effectively amplifying disinformation campaigns. The attack vector does not target the AI models' core infrastructure but rather exploits the way these systems crawl and index web content. Security analysts note that the campaign leverages search engine optimization techniques to ensure malicious pages are prioritized by AI crawlers. The manipulated responses may direct users to phishing sites or content designed to spread false narratives. OpenAI and Google have been notified, and both companies are investigating the reports. As of now, no patches or specific mitigation steps have been released, though both vendors are reportedly working on improved detection and filtering mechanisms.