'GhostJacking' Exposes Identity Governance Gaps in AI Agents
2026-08-11T14:51:33Z•21334a13e8e9580dc4e27faebbffe6e316c13fba19a088407ade529c8799eb8e
CVE-2026-18577AI agentsAI securityCVEICS/OTMetabaseN-ablePLCRMMagent hijackingauthentication bypasscritical infrastructuredata exposuredeepfakesdevice-code phishingidentity governancemalwaremobile RATphishingprompt injectionsandbox escapesocial engineeringsupply chain securityvishingwater utilitieszero-day
What happened
Dark Reading feed covering emerging AI-agent security threats, prompt injection and sandbox escapes, critical infrastructure attacks, actively exploited vulnerabilities, phishing and social engineering, malware, data exposure, and cybersecurity policy. The most severe items include a maximum-severity Metabase SQL zero-day without a CVE, attacks against Internet-exposed water-system PLCs, and an exploited N-able authentication-bypass vulnerability (CVE-2026-18577).
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- darkreading
- Record identifier
- 21334a13e8e9580dc4e27faebbffe6e316c13fba19a088407ade529c8799eb8e
- Enrichment time
- 2026-08-11T14:51:33Z
- AI-assisted enrichment
- Yes
This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.