Google rapidly eliminated the defined problems, but researchers emphasize that the event shows risks associated with the security deficits of AI, which can be used for attacks and as a tool for them.
Gemini Cloud Assist, the first of the security vulnerabilities, the attackers through the HTTP user agent, then allowed to bring harmful commands displayed in the diaries. When users returned to the system with the desire to analyze these diaries, artificial intelligence processed secret instructions.
The second problem was about the search system in which the attackers could bring malicious demands to the past of the browser, which AI perceived as legitimate teams.
The third vulnerability, which is considered to be the most dangerous, affected the Geminini browser vehicle. The attackers can force AI to pass the harmful connections that lead to leaks by transferring user data to URL parameters.
The attacks were carried out in two stages: first, cyber criminals entered secret commands or malicious requests, and then released the data in an unendable way. Researchers have shown that such actions can compromise on user preferences, geographical location and other secret information.
Source: Ferra
I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.
