Gemini vulnerability allows you to use a brief content of letters in gmail for phishing

Vlad Cherevko I have been interested in all kinds of electronics and technologies since 2004. I like to play computer games, and I understand the work of different gadgets. I regularly monitor the news of the technology in the world and write materials about it.

Gmail -integrated gemini has a critical vulnerability that allows hackers to sell phishing attacks through artificially generated letters. About it writes the portal BleepiingComputer with reference to 0din.

The vulnerability was discovered by Marco Figeroa – Genai Bug Bounty manager in Mozilla. According to Figeroa, malefactors can hide instructions in the body of an e -mail, formatting them in white and reducing the font to zero, which makes the text invisible to humans, but accessible for Gemini analysis. As a result, the AI can automatically add false warnings to the resume, such as breaking the password, along with the fake support number.

Although some users do not respond to such messages, others can be trapped because of the emotional impact of such content. Figeroa emphasizes that security teams can develop methods for detecting hidden information, as well as analyze the results that generates the AI for URL, phone numbers or urgent messages.

BleepingComputer turned to Google on this vulnerability in Gemini. The company representative replied that he had not seen any evidence of abuse, but added that Google was already working on protection and would soon introduce additional safety measures.