←Story detail

Trending story

Noam Brown says they are using an agent-team-like approach for red teaming

1 reports1 story updatesUpdated Oct 24 04:28

Understand this story

AI overview

Noam Brown stated that the team is using an agent-team-like approach for red teaming to find and fix infrastructure vulnerabilities. He mentioned that last month a large agent team consumed over 100 billion tokens to solve the Navier-Stokes Millennium Prize Problem, and now the team is applying a similar approach to red teaming. He cited an Eric Wallace post saying that the team built a large team composed of the models with the strongest cyber offense and defense capabilities, collaborating in an offline training environment to find zero-day vulnerabilities; although no path bypassing CoT monitoring to reach the internet was found, they discovered methods to escape mainstream virtual machines, exploit network infrastructure, and communicate across virtual machines via hardware side channels, and the related findings were used to harden infrastructure and submit a large number of patches upstream to open-source software packages. The currently verifiable information comes from this quotation, with no further independent sources or follow-up developments yet.

AI-generated from public reports

Latest development · Report

Noam Brown: Using large-scale agent teams to red-team infrastructure vulnerabilities

Developments

1 confirmed developments

  1. Noam Brown said they are using an agent-team-like approach for red teaming

    Source:Noam Brown

Reports and discussion timeline

Browse reports and discussion by time. Their evidence levels remain distinct.

Oct 24

  1. Noam BrownSelected

    Noam Brown: Using large-scale agent teams to red-team infrastructure vulnerabilities

    Noam Brown stated that last month a large agent team consumed over 100 billion tokens to solve the Navier-Stokes Millennium Prize Problem, and the team is now using a similar approach for red-team testing to find and fix infrastructure vulnerabilities. He cited an Eric Wallace post saying the team built a large team composed of the models with the strongest cyber offense and defense capabilities, collaborating in an offline training environment to find zero-day vulnerabilities; although they did not find a path to bypass CoT monitoring to reach the internet, they did find methods to escape mainstream virtual machines, exploit network infrastructure, and communicate across virtual machines via hardware side channels, and the related findings were used to harden infrastructure and submit a large number of patches upstream to open-source software packages.

Related stories

  1. Related storiesNSA warns that AI discovered two critical PHP vulnerabilities
  2. Related storiesAnthropic unveils proactive AI agent Claude Duet
  3. Related storiesNVIDIA launches DSX POD modular AI factory reference design
  4. Related storiesMicrosoft says China-linked hackers breached about 2,250 organizations using AI agents