Threat Actors are Finding Their Way into Your AI Summaries

How AI summary accuracy (and inaccuracy) has changed

AI summaries have become ubiquitous in the modern search engine. They are often the first result that users see and helpfully contain links alongside their authoritatively worded answers. However, these answer engines are fallible, and their responses could include malicious links to phishing sites, malware, and more.

Netcraft tested queries about the login pages of several of the world's largest banks and retail brands in ChatGPT, Copilot, Gemini, and Perplexity. Those queries generated 2,905 responses containing a total of 20…