International
The Morning Dev

Only the news worth knowing

‹
 
B46/100

Incident · high

Google reports AI agent turned security patch into working exploit

Tier B · Patch this week – Put it in the sprint

Who’s exposed: Organizations slow to deploy newly released security patches

Tier B because: Exploited in the wild OR public exploit

Reach

Attack

Damage

Fix

Exploited in the wild · Secrets exposed

The Brief

  • Google's Threat Intelligence Group (GTIG) reports a suspected Russian-speaking operator used an AI agent to turn security patches into working exploits with no human in the loop.
  • The operator then stole data from victim organizations, according to the report.
  • The report also describes attackers using autonomous multi-agent offensive frameworks such as ARTEX, CAIRN, CyberStrikeAI and PentAGI.

Why it matters

Patches can now be turned into working exploits faster and without human effort, so the time between a patch release and its deployment matters more.

The checklist behind the tier

Checklist · 3 of 10 yes
YesHits tech teamsJudgment: exploits target patched enterprise software and servers that IT and ops teams run.
NoCore infrastructureJudgment: no core infrastructure software such as OpenSSH, the Linux kernel or GitHub is named as compromised.
NoPopular toolJudgment: no widely used framework, tool or package is named as the flawed or compromised product.
YesExploited in the wildThe Register: operator used an AI agent to make working exploits before stealing data from victim organizations.
NoPublic exploitSources describe attacker use of exploits; none say exploit code or a PoC was published.
NoRemote, no loginNo source states whether the exploits work unauthenticated over the network.
NoFull takeoverNo source reports code execution, root or admin rights; sources only mention data theft.
YesSecrets exposedSecurity Affairs: GTIG warns AI agents automate exploits and steal credentials; The Register: data stolen from victims.
NoSupply chainNo source reports a compromised package, update, build pipeline or signing key.
NoNo fix yetThe attacks used security patches, meaning fixes were already released; no source says a fix is missing.

Timeline

  1. Google GTIG and Mandiant publish report on AI agent turning a security patch into a working exploit

  2. The Hacker News

    Google Finds Hackers Using AI Agents to Build and Run Exploits From Software Patches

  3. Cloud Blog

    GTIG AI Threat Tracker: Agentic Threat Activity Crosses a Significant Threshold

  4. BleepingComputer

    Google warns that fully-automated AI attacks are coming soon

  5. www.theregister.com - Articles

    Google: AI agent turned a security patch into a working exploit with no human help

  6. Techmeme

    Google's TIG reports the first observed instance of attackers using an autonomous AI framework with no human in the loop to turn security patches into exploits (Duncan Riley/SiliconANGLE)

  7. The Decoder

    Google warns that fully automated cyberattacks have arrived and will get worse fast

  8. Security Affairs

    Google GTIG Warns AI Agents Automate Exploits, Steal Credentials and Scale Cybercrime

  9. The New Stack

    Security patches are supposed to protect you. Google found AI agents turning them into exploits.

  10. Help Net Security

    Google finds AI agent writing and launching exploits with no human in the loop

  11. DEV Community: ai

    Google detecta por primera vez un agente de IA que convierte parches de seguridad en exploits sin intervención humana

Full coverage · 10 sources

First seen · 10 sources · Full story page · How tiers are decided