While Anthropic claims its Claude Opus 4.6 can barely find zero-days, Mythos Preview can pop up working exploits 72.4 percent of the time. It's a good thing Anthropic has limited its use for now; if ...

Every conversation you have with an AI — every decision, every debugging session, every architecture debate — disappears when ...

AI firm Anthropic accidentally leaked its Claude Code source code via an npm package, revealing unreleased features like an ...

Although executed by different attackers – Axios by North Korean-linked goons, and Trivy et al. by a loosely knit band of ...

Google has launched TorchTPU, an engineering stack enabling PyTorch workloads to run natively on TPU infrastructure for ...

A Python package presented as a privacy-first shortcut to AI models has been unmasked as a supply-chain threat that quietly captures user prompts, leans on a private university service without ...

Mark Collier briefed me on two updates under embargo at KubeCon Europe 2026 last month: Helion, which opens up GPU kernel ...

Over 1,000 exposed ComfyUI instances exploited via unauthenticated code execution, enabling Monero mining and botnet expansion.

Anthropic just built an AI model so dangerous it had to cancel the public launch. During pre-deployment testing, the company’s newest frontier model, Claude Mythos Preview, proved so adept at hunting ...

Aethyr Research has released post-quantum encrypted IoT edge node firmware for ESP32-S3 targets that boots in 2.1 seconds and ...

PM This week in cybersecurity: botnets, RCE flaws, AI-driven attacks, stealers, and more. Fast, no-fluff roundup.

The open-source project maps directly to OWASP’s top 10 agentic AI threats, aiming to curb issues like prompt injection, ...