The crazy, true story behind the first AI-powered ransomware Summary A team of NYU engineers developed a proof-of-concept they call Ransomware 3.0 to test whether large language models (LLMs) can…
Shell to pay: Crims invade your PC with CastleRAT malware, now in C and Python Summary A criminal crew known as TAG-150 has released two variants of CastleRAT — one…
Summary Researchers at Nanjing University and the University of Sydney have developed A2, an AI-powered agent that mimics human vulnerability discovery for Android apps and automatically validates flaws with proof-of-concept…
Proof-of-Concept in 15 Minutes? AI Turbocharges Exploitation Summary An Israeli researcher duo built an automated offensive research pipeline, dubbed Auto Exploit, that uses prompts to a large language model (Anthropic’s…
Boffins Build Automated Android Bug Hunting System Summary Researchers at Nanjing University and the University of Sydney have built A2, an AI-powered agent that automates discovery and validation of vulnerabilities…
Proof-of-Concept in 15 Minutes? AI Turbocharges Exploitation Summary Two independent Israeli researchers built an AI-powered offensive research pipeline, dubbed Auto Exploit, that uses LLM prompts (Anthropic’s Claude-sonnet-4.0), CVE advisories and…
Boffins Build Automated Android Bug Hunting System Summary Researchers from Nanjing University and the University of Sydney have built A2, an AI-powered agent that automates discovery and validation of Android…
Proof-of-Concept in 15 Minutes? AI Turbocharges Exploitation Summary An AI-driven offensive research pipeline developed by two Israeli security researchers — dubbed Auto Exploit — used prompts to a large language…
Boffins Build Automated Android Bug Hunting System Summary Researchers from Nanjing University and the University of Sydney have created an AI agent, dubbed A2, that automates discovery and validation of…
Proof-of-Concept in 15 Minutes? AI Turbocharges Exploitation Summary An AI-driven offensive research pipeline developed by two independent Israeli researchers — dubbed Auto Exploit — used prompts, CVE advisories and open-source…
