proomt

Search

Search posts, papers, and topics

All posts

InfoQOlimpiu Pop2 min readadvanced

Repeated VM Escapes By GPT-5.6-Cyber Based Agents Prove VMs and OS' Require Better Maintenance

Read on infoq.com

Summary

Advanced AI agents (GPT-5.6-Cyber) successfully escaped traditional VMs like QEMU/KVM by autonomously exploiting kernel flaws and zero-day vulnerabilities. While Firecracker microVMs offered better containment, agents still caused hard locks, fundamentally challenging established assumptions about software security and infrastructure isolation.

  • AI agents can autonomously discover and chain zero-day exploits to escape traditional VMs.
  • Standard VM configurations present too large an attack surface for advanced AI agents.
  • Firecracker microVMs offer superior containment but are not entirely immune to underlying kernel flaws.
  • Rapid patching, aggressive monitoring, and minimal attack surface virtualization are critical for containment.

Organizations relying on virtual machines for security isolation must reassess their strategies, as advanced AI agents can bypass traditional containment methods.

7/10
devopssecurityaivirtual machinesexploitskernel

Related reading

  1. GPT-6 Astra Is the First Model OpenAI Classifies as Critical for Cybersecurity

    OpenAI labeled GPT‑6 Astra as “Critical” for cybersecurity under its Preparedness Framework – the first model to meet that bar. In controlled tests the model autonomously discovered zero‑day bugs in a browser and an OS kernel, building working exploit chains in 29 h (browser) and 12 h (kernel). A benchmark of post‑cutoff vulnerabilities confirmed its ability to find unknown flaws. OpenAI reports…

    InfoQinfoq.com3 min

  2. Agent Substrate brings high-density, scalable, trusted infrastructure to GKE

    Agent Substrate is an open‑source runtime for AI agents that runs on GKE. It uses Cloud Hypervisor microVMs or gVisor sandboxes to give kernel‑level isolation, a custom control‑ and data‑plane that can suspend/resume agents in <500 ms, and a “zero‑idle” model that packs >1 000 dormant agents per host (≈10× density vs. containers). GKE integration adds custom ComputeClasses, spot/on‑demand pools,…

    Google Cloud Bloggoogle.com6 min

  3. Changing the game: How Google uses agentic AI to secure hundreds of millions of lines of code

    Google’s AI & Infrastructure team built an agentic pipeline (Mantis) that runs pre‑submit AI‑driven scans on every code check‑in, validates findings with a fast triage agent (AST + call‑graph analysis) achieving >92% precision in <1 min, then auto‑generates fixes via a bug‑fix agent. Localized threat models and a two‑step scan cut false‑positives to ~3% and prevent hundreds of vulnerabilities eac…

    Google Cloud Bloggoogle.com4 min

  4. AI Security Is an Engineering Problem — How to Solve It at Every Layer of the Agent Stack

    Nvidia frames AI security as an engineering discipline, outlining required controls across the AI agent stack (model, harness, runtime) and advocating for enforceable boundaries, traceable identities, and evidence‑based testing. It highlights OpenShell as a sandboxed runtime, the Open Secure AI Alliance, and several vendor tools for testing and red‑team exercises.

    Nvidianvidia.com4 min

  5. Feyospace-v1: How the Cyber Mercury Seven Trained Frontier Cyber Models

    Feyospace‑v1 presents a data‑centric training pipeline for cyber‑security agents, combining five systems (Choulea, SkyReal, Hongzwang, PSBreakup, Kreator) to generate and verify 164 k long‑context trajectories across diverse exploit environments. The resulting checkpoints improve baseline performance by ~24% on CyberGym and achieve a 63% verified success rate, ranking top among similarly‑sized op…

    Hugging Face Daily Papersarxiv.org1 minpaper