Black Hat 2026: Open-source tool makes red teaming AI agents up to 125x cheaper
NVIDIA researchers shared the “AgentBreaker” tool and how they tuned an open-source model to perform at frontier level.
SC Media
20 articles
Frank Vukovits on how Delinea delivers runtime authorization for AI agents
Vukovits breaks down why verifying access at connection time is no longer enough.
Intruder's Chris Wallis on AI pentesting and the future of cybersecurity
Wallis dives into the nuances of AI-enabled security and how these offerings will impact midmarket security teams.
Ido Geffen on why Novee owns the full AI pentesting stack
Novee's Geffen breaks down what it means to own the full AI pentesting stack.
Ramin Farassat on how Menlo Security is securing AI agents from prompt injection
Farassat discusses how security teams can put controls on the agent attack surface without blocking agentic AI outright.
North Korea linked to new NullReceiver C2 technique
The Hacker News reports that North Korean threat actors have developed a new command-and-control (C2) technique called NullReceiver, an evolution of the Ethe...
OVSwrap vulnerability allows local privilege escalation on Linux
As reported by Security Affairs, a 13-year-old Linux kernel flaw named OVSwrap was disclosed, enabling local users to gain root privileges on many default-co...
When AI Commits Felonies - PSW #938
Mimecast's Leslie Nielsen on why human risk doesn't stop at people
Mimecast's Nielsen argues that treating agentic AI as a new security category is the wrong instinct.
Optiv's John Hurley on simplifying cybersecurity
John Hurley discusses Optiv's ability to advise, deploy and operate complete cybersecurity programs.
Paperclip authorization bug exploited, leads to control plane takeover
Critical Paperclip flaws let attackers seize AI agent platforms, escalate privileges, and execute code.
Google's John Hultquist on outpacing the adversary with AI threat defense
John Hultquist discusses Google’s approach to combating today’s threats.
Apiiro's Idan Plotnik on AI writing code that sharpens attacks
Apiiro's Idan Plotnik makes the case for a prevention-first model.