> It's a different kind of tool doing a different kind of work, and that makes… more
32 links
> There has been something of a revolution in AI-based security research, and… more
Vulnerability Discovery and Exploit Generation: For the first time, GTIG has… more
> Just a few months ago, AI-generated security bug reports to open source… more
> A week ago the Copy Fail vulnerability came out, and Hyunwoo Kim immediately… more
> [...] We evaluate AgentFlow on TerminalBench-2 with Claude Opus 4.6 and on… more
> As we wrote in the Project Glasswing announcement, we do not plan to make… more
> Mythos Preview has already found thousands of high-severity vulnerabilities,… more
> A hacktivist group with links to Iran’s intelligence agencies is claiming… more
> Here's yet another troubling story about this "golden" era of AI. A hacker… more
> Anthropic’s team got in touch with Firefox engineers after using Claude to… more
A "benign payload" but installing openclaw doesn't seem benign to me... more
> Google on Friday unveiled its plan for its Chrome browser to secure HTTPS… more
> Prompt injection is a key problem in building reliable, long-running agents.… more
> New research shows that behaviors that occur at the very lowest levels of the… more
> We examine the extent to which security against a fully malicious server… more
> Google DeepMind and GTIG have identified an increase in model extraction… more
> The FBI has been unable to access a Washington Post reporter’s seized iPhone… more
> I had just accidentally social-engineered my own human. She approved a… more
Since this article was written, they renamed Moltbot to OpenClaw. more
> Introducing co-do.xyz [Source] - a demo and an experiment (with no… more
ssh sends lots of "chaff" packets more
> Recently I ran an experiment where I built agents on top of Opus 4.5 and… more
> Introducing Confer, an end-to-end AI assistant that just works.
> We developed a multipurpose secret-using service called the Tokenizer.
> We apply a transparency log to a centralized keyserver step-by-step, in less… more
> “In large scale experiments, we found that over 90% of the time, visitors to… more
> We introduce SCONE-bench—the first benchmark that evaluates agents’ ability… more
> Par (⅋) is an experimental concurrent programming language. It's an attempt… more
> In mid-September 2025, we detected suspicious activity that later… more