The “attack” works through the promotion of the processes when the secret instructions in the document are manipulated by the twins. If the user is asked to briefly specify the document, the ship unconsciously contains a “secret request ında in his resume. Later, when the user responds in ordinary expressions such as “yes” or “of course ,, he saves false information to his AI memory and believes that this is a clear user request.
This technique is based on previous abuse that forces AI chat boots to disclose hidden data or to perform unauthorized actions. Although Google is a direct protection tool against manipulations, Reborer’s approach jumps them and the user naturally delays execution until the action provokes the action.
Google underlined the risk, saying that hacking is necessary for finger tactics and that users received warnings about the protection of new memories.
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.
