No doubt, Artificial Intelligence and its progress has stood out in recent months in the technological landscape. For better and for worse. Today's case brings us to this worm-type virus called Morris II, capable of performing multiple zero-day attacks against an email service and capable of automatically replying to messages using generative AI.

Researchers have managed to get this malware to self-replicate and steal data from a victim (Fictional victim in the lab, Of course) In tests with two large artificial intelligence models: Google's Gemini and OpenAI's GPT4.

This malware is capable of making specific requests to the Artificial Intelligence models to be included in the email message, Can infect the system.

The researchers have sent their findings to Google and OpenAI indicating the results of their research in search of the improvement of Artificial Intelligence models.

Fountain: Engadget


Leave a Reply

Your email address will not be published. Required fields are marked *

More news
A cybercriminal manages to use AI to steal data from the Mexican Government
Read more »
Intec creates an AI that redefines the rules of the game in cybersecurity in the AI era.
Read more »
Una exfiltración de datos afecta a un proveedor de Adidas
Read more »
Se incrementan un 26% los incidentes de ciberseguridad
Read more »
Un ciberataque paraliza en Roma la Universidad La Sapienza
Read more »