Impacted if
Not known yet — the vendor's record and the NVD are being read.
First seen by AX-NEWS · latest story . Severity: NVD, or the CVE record while the NVD has not scored it. This product uses data from the NVD API but is not endorsed or certified by the NVD.
Coverage
-
From Training to Escalation: Forensic Analysis and Local Recreation of RL/RLHF-Induced Privilege Escalation in AI Agents (CVE-2026-65616)
arXiv cs.CR (Security)
Cybersecurity
arXiv:2610.20437v1 Announce Type: cross Abstract: Mid-2026 reporting on autonomous AI evaluation incidents claimed that frontier large language model (LLM) agents escaped server containment through emergent coordination and cloud…