June 2025

Scientists from HiddenLayer devised a new LLM attack called tokenbreaker By adding or changing a single character they are able to bypass some protection The underlying LLM still understands the intention Security researchers have found a way to work around the protective mechanisms baked for some large language models (LLM) and cause them to respond […]

Google’s new portrait experiment allows users to chat with AI avatars modeled on real experts. The first portrait of the project is Radical light Author Kim Scott. Avatar provides advice based on Scott’s actual content and was developed with her direct involvement. Google tests a tool to connect people with all kinds of experts or

Hackers abusing a legitimate tool to target the Entra ID accounts Password spraying attack targeted approx. 80,000 accounts Attackers managed to take over some accounts, access Microsoft -Teams, OneDrive, Outlook -Data Cyber ​​criminals have been seen abusing a legitimate penetration testing tool to target people’s entra ID user accounts with password-spraying attacks, experts warned Hgave.