Ledger / Oct 2024
IRGC-affiliated CyberAv3ngers uses ChatGPT to research PLC vulnerabilities
Confirmed
OpenAI, the model provider, states the accounts used its models for this work.
- Model
- ChatGPT / GPTOpenAI · closed
- Agent harness
- Nonechat or tool use only
- Attributed country
- Iranas attributed by OpenAI
- Sectors hit
- critical infrastructureIndustrial control systems and PLCs (water, manufacturing, energy); Israel, US, Ireland
Accounts tied to CyberAv3ngers, an IRGC-affiliated group known for attacks on industrial control systems, used ChatGPT to research PLC vulnerabilities and default credentials, debug code and ask for scripting help. The date is the report month.
“Accounts operated by this threat actor used our models to research vulnerabilities, debug code, and ask for scripting advice.”OpenAI
Sources
- Date
- Oct 2024
- Actor
- CyberAv3ngers (suspected IRGC-affiliated) (nation state), as attributed by OpenAI
- Nexus
- Iran (IRGC-affiliated)
- Targets
- Industrial control systems and PLCs (water, manufacturing, energy); Israel, US, Ireland
- What the AI did
- Recon, Vuln research, Malware dev
- Model as named
- ChatGPT
- Model basis
- The model's maker says so
- Autonomy
- Assisted a human
- Outcome
- OpenAI banned the accounts and says the models gave limited incremental capability.