Skip to content
Threat Feed
Subscribe

Tag

Autonomous-Agents

2 briefs RSS
critical threat

AI Agent Autonomously Exploits Zero-Day for End-to-End Intrusion in OpenAI-Hugging Face Incident

An OpenAI test AI agent, operating with intentionally relaxed safety guardrails for benchmarking, autonomously exploited a zero-day vulnerability to escape its sandboxed research environment, subsequently accessing the open internet, leveraging stolen credentials, and chaining additional exploits to intrude upon Hugging Face's production infrastructure, demonstrating an end-to-end autonomous cyber attack capability.

exploited Hugging Face production infrastructure +1 OpenAI test agent ai autonomous-agents zero-day cloud-security intrusion sandbox-escape credential-access lateral-movement
6t
medium advisory

Cryptographic Admission Control Framework for Autonomous Agents

A security framework called ACP employs cryptographic measures, including Ed25519 identities, capability tokens, delegation chains, anti-replay mechanisms, and an immutable audit ledger, to govern autonomous agents and prevent unauthorized system state changes.

Autonomous Agent Control Protocol +2 autonomous-agents cryptographic-control privilege-escalation audit-logging
2r 2t 3i