Single Line of Code Can Jailbreak 11 AI models Including ChatGPT, Claude, and Ge...
A newly detailed jailbreak technique known as “sockpuppeting” allows attackers to bypass the safety guardrails of 11 maj...
Found 177 results
A newly detailed jailbreak technique known as “sockpuppeting” allows attackers to bypass the safety guardrails of 11 maj...
A single threat actor compromised nine Mexican government agencies and stole hundreds of millions of citizen records in...
OpenAI has disclosed a security incident tied to the compromise of Axios, a widely used third-party JavaScript developer...
The cybersecurity community is on high alert after the disclosure of a critical security flaw in Axios, a widely used pr...
A widespread cyber espionage campaign leveraging 108 malicious Google Chrome extensions. According to a recent report by...
OpenAI has unveiled GPT-5.4-Cyber, a specialized variant of its flagship GPT-5.4 model fine-tuned for advanced defensive...
A relatively unknown ransomware group called Payouts King has emerged as a serious cybersecurity threat, carrying the to...
A North Korean state-sponsored threat group is running an active campaign that tricks software developers into installin...
North Korea has been running one of the most quietly effective cyber fraud operations in recent years. State-sponsored o...
OpenAI has announced a new Bio Bug Bounty program for GPT-5.5 as part of its efforts to improve safety controls for adva...
OpenAI has published a comprehensive cybersecurity action plan titled “Cybersecurity in the Intelligence Age: An Action...
A large-scale phishing campaign has been quietly targeting organizations across the United States, using fake event invi...