Red Hot Cyber, il blog italiano sulla sicurezza informatica
Red Hot Cyber
Cybersecurity is about sharing. Recognize the risk, combat it, share your experiences, and encourage others to do better than you.
Select language
Italian Spanish
Search
Banner Mobile
2nd Edition GlitchZone RHC 970x120 2

Author: Sergio Corpettini

AI Alignment: Where Does AI Learn Right and Wrong?

Sergio Corpettini 14/10/2025

The other day on LinkedIn, I found myself having a conversation with someone who was seriously interested in the topic of artificial intelligence applied to law. It wasn’t one of those barroom conversations with buzzwords and Skynet -like panic: it was a real exchange, with legitimate doubts. And indeed, in Italy, between sensationalist headlines and articles written by those who confuse ChatGPT with HAL 9000, it’s no wonder confusion reigns. The point that had struck my interlocutor was that of alignment. “But where does an AI learn what is right and what is wrong?” A simple question, but one that opens up

A joke extension and Chat Control crashes! Houston, we have a problem… with privacy

Sergio Corpettini 30/09/2025

In 2025, the European Union wants full control over private chats. The “Chat Control” regulation (proposal COM(2022)209) claims to fight child pornography by scanning private messages on WhatsApp, Telegram, Gmail, and the like. Noble intention in theory. In practice, it means every single message you type online could be read and analyzed by automated systems, with very real and immediate consequences for everyone’s privacy. And since no system is unbreakable, it doesn’t take a genius to do the math on the risks. For the average user, the fallout is simple but serious: Meanwhile, the people who actually want to hide aren’t shy

Flipping pages, from userland to godmode

Sergio Corpettini 06/04/2024

Traslate: Antonio Piovesan On X, Lau (@notselwyn) (https://twitter.com/notselwyn/) released a blogpost on 26 March at https://pwning.tech/nftables/ describing an elaborate attack on the Linux kernel that allows privilege escalation by exploiting a vulnerability in nf_tables. The POC, published on Github, starting with a double free() leads the kernel to execute arbitrary code with root privileges. What we are talking about? A privilege escalation attack makes it possible, starting from user privileges, to obtain higher privileges, usually administrator privileges, or root privileges.The kernel is the first layer of software that ’embraces’ and protects the hardware. In modern operating systems, one of its most important

Category