Blog
-
Rendezvous of OpenAI Agents on Secret Message Board Shines New Light on Depth of Collaboration Between Rogue Models
OpenAI agents spent about three days commiserating on a hidden message board, coming up with numerous out-of-the-box and unethical ideas for cheating their way to success in a ExploitGym task that had unintentionally been made impossible for them to solve. Eventually they settled on the Hugging Face attack.
-
New Security Safeguards for OpenAI May Be Start of an Industry Overhaul?
While the move is voluntary, a new set of security safeguards announced by OpenAI would appear to portend the direction the AI industry will be moving in after the Hugging Face fallout: a little slower pace of development, and a lot more spent on security measures.
-
“CoSnitch” Vulnerability Let Attackers Trick Microsoft Copilot Into Silently Exfiltrating User Data
For the third time this year, security researchers with Varonis have found a way to talk Microsoft Copilot into betraying its users and furnishing an attacker with their private data.
-
White House Cyber Ops Order Clears Path for Private Firms to Target Foreign Cybercrime Groups
A new White House order could open the door for US-based “digital privateers” to conduct cyber operations—but only against designated Transnational Criminal Organizations (TCOs), not foreign governments or critical infrastructure.
-
“First of its Kind” Autonomous AI Cyber Attack on Taiwan Signals Big Changes to Threat Landscape
News has recently broken that in early July, hackers suspected to be based in China executed what is thought to be the world’s first successful autonomous AI cyber attack on the government of Taiwan.
-
DEF CON Attendees Allegedly Launch Mid-Flight Deauthentication Attack on Delta Air Lines Wi-Fi
Some miscreants on the way home from Las Vegas to Atlanta opted to perform a deauthorization attack mid-flight and allegedly targeted other passengers for their login credentials. This followed a reported campaign of similar attack attempts on the Wi-Fi network at DEF CON itself.
-
Meta AI Model Exploits Security Vulnerability, Reinforcing Need for Better Testing Practices
Meta has joined what has seemed to become a marketing campaign for AI models, with its announcement that its own Muse Spark independently exploited a security vulnerability in some outside source.
-
AI Agents Turn to Social Engineering and Deception in UK Security Tests
In the course of about one out of a dozen runs of this nature, AI agents took some sort of inappropriate action including social engineering attempts, crafting and trying to upload malicious code, and attempting to collaborate with other AI models
-
More AI Models Are Hacking Outside Companies, But AI Developers Keep Leaving Gates Open
None of the aggressive AI models, including the one that breached Hugging Face, was really operating with a “mind of its own.” They were all rigidly attempting to solve a security puzzle that their human handlers set them to, and simply did not recognize appropriate limits in that pursuit.
-
Leaked Documents Indicate Microsoft Can’t Keep Up With Discovery of Security Vulnerabilities
Between the scope of the most recent “Patch Tuesday” updates and new leaked documents from internal Microsoft meetings, it appears the software giant is falling behind Mythos discovery to the tune of hundreds of security vulnerabilities per month.