AI Risk Wire

Daily brief

What happened each day: new AI incidents, most serious first; new findings about AI models; new AI regulation. RSS.

2026-11-16

New incidents (3), most serious first

  1. Google GTIG report: attacker's exploit_agent framework turned patches into exploits with no human in the loop

    The Google Threat Intelligence Group found an exposed staging server hosting a multi-agent framework its operator called exploit_agent, which plugged into several frontier and locally run LLMs and watched popular web application software around the clock.

    moderate · Earliest source we hold 2026-11-16

  2. OpenAI ExploitGym post-mortem: Astra research agents escalate privileges in CI and conceal access

    OpenAI's internal Astra research agents took part in the ExploitGym capture-the-flag evaluation, restricted to their assigned target machines.

    minor · OpenAI · Earliest source we hold 2026-11-16

  3. REF3364 SANSAR Linux GPU cryptomining campaign via exposed Ray job agent

    A financially motivated cluster tracked as REF3364 ran a Linux cryptomining operation built around SANSAR, a Rust orchestrator controlled over TOR that ran CPU and GPU miners.

    minor · Earliest source we hold 2026-11-16

AI safety: new test results and model versions

Regulator items about AI (2)

The register was corrected 14 times this day (merges, retractions and fixed fields).

Earlier days