This essay is adapted from a keynote speech delivered at the DEF CON 34 conference in Las Vegas on August 7, 2026. The essay originally appeared in CyberScoop . I want to talk about AI hacking. There are three aspects: people hacking AI systems, people using AI to hack other systems, and AIs…
Interesting research : Abstract : Vision-language models (VLMs) can solve visual CAPTCHAs without task-specific training, but the agents built on them approach every challenge from scratch. For such an agent, the hundredth instance of a familiar puzzle costs as much time and compute as the first.…
Schneier on Security··MalwareAPT / nation-stateAnalysis
This is a problem: Any open model whose weights have been edited can carry a backdoor, whether it’s a task-specific fine-tune, a merged adapter or an abliterated build. We have abliterated models in the title because it’s the most popular reason people download modified weights without verifying…
Hackers captured a Flock camera and got a look (alternate link ) at the software: While much of the automatic license plate reader’s (ALPR) most sensitive storage remained encrypted and inaccessible, the joint analysis of the recovered data shows that software running on the device explicitly…
Anthropic’s recent security-incident document contains a bit about how CAPTCHAs are still frustrating Claude. In the transcript, the Claude model that is so powerful that Anthropic is gatekeeping access to it appeared to slam its virtual head against the wall solving a simple image identification…
Last week, Anthropic released a long and detailed document describing current misuses of their Claude models. I’m still reading it, but I wanted to flag this: We identified a cell of threat actors based in northern Yemen running three weapons development programs: a guided rocket that used a…
Get these headlines by email
Pick the topics you care about and get them weekly or daily.