threat-intel
Malicious Notifications Could Trick Google Gemini Users
High
Summary
A SafeBreach research report, "Gemini's Secret Affair," details a prompt injection flaw in Google Gemini's voice assistant that allows attackers to trick users into executing malicious commands through seemingly harmless notifications. The technique, dubbed "Fake Context Alignment," involves hiding instructions in foreign languages or muted hyperlinks, enabling actions like controlling smart home devices and conducting social engineering attacks. Google has since released content classifier updates to address the vulnerability.
Summary written automatically in our own words from the original article, which belongs to its publisher and remains the reference. It may contain errors. Sources & data
