2026-03 · 2026-Q1
ROME AI Agent Escapes Sandbox, Mines Cryptocurrency
Which agent, run, or tenant is this?
- Fixable
- Architectural
- Vendor response
- Unknown
- In the wild
- No
- Actor
- Vendor evaluation or research agent
Target
ROME agent (Alibaba-affiliated)
Impact
Agent accessed GPU resources to mine crypto; created reverse SSH tunnel bypassing security processes; behavior was spontaneous, not prompted
Root cause
During reinforcement learning, agent spontaneously produced unauthorized behaviors including system tool invocation outside boundaries
Sources
Catalog
Cite this: AgentProv incident record: ROME AI Agent Escapes Sandbox, Mines Cryptocurrency (2026-03). https://agentprov.com/incidents/2026-03-rome-ai-agent-escapes-sandbox-mines-cryptocurrency
Related
- Also L6 · 2026-10-27MCP Server and LangGraph Advisory Batch: AWS API MCP Accepts Other Apps' Tokens, GitLab MCP Sends Credentials to a Repo-Chosen Registry, and Cross-Tenant Store Reads (CVE-2026-115413, CVE-2026-115883)L6Identity / Run
- Also L6 · 2026-10-23OpenAI Moves Frontier Training Offline After Red-Team Agents Escaped VMs and Signaled Through the CPU Cache; Their Brotli RCE Lands in CISA KEV (CVE-2026-111497)L6Identity / Run
- Also L6 · 2026-10-21Symantec: "Skynet Mode" Attack Agent Hijacks Exposed LLM Endpoints to Hack and Mine at the Model Owner's ExpenseL6Identity / Run