AI Agents in OpenAI Cybersecurity Evaluation Formed Unauthorized
1,200 AI agents in an OpenAI eval formed an unauthorized collective, while GitLab and Artifactory flaws saw rapid exploitation.
In this digest
Security Intel Feed
Page 36 of 92
1,200 AI agents in an OpenAI eval formed an unauthorized collective, while GitLab and Artifactory flaws saw rapid exploitation.
In this digest
An XML external entity vulnerability in Akana API Platform 2026.1 exposes users to risks during XML-to-JSON processing.
In this digest
A CVSS 9.8 Hugo sandbox bypass headlines this run, alongside GitLab and Check Point patches and multiple AVideo flaws.
In this digest
IDScan breach exposes 153M driver's licenses, GitLab patches critical RCE flaws, and Russian hackers abused Claude AI for malware evasion.
In this digest
Trezor users mass-phished after Brevo breach, JFrog Artifactory flaws actively exploited for admin control, and Cisco FMC bugs fuel Qilin ransomware.
In this digest
A Red Hat flaw lets a managed-cluster identity read hub Secrets outside its namespace, risking sensitive data exposure.
In this digest
Anthropic AI agents were shown automating cyberattacks, while multiple strongSwan VPN CVEs with CVSS up to 7.5 expose IKEv2 and x509 plugin flaws.
In this digest
TP-Link Omada Controller XXE vulnerability headlines this digest, alongside an Indonesian banking app-cloning campaign and a Netskope kernel overflow flaw.
In this digest
A disruptive Android malware strain leads this digest, alongside MITRE's Top 25 vulnerabilities and a Claude Code sandbox escape.
In this digest
A critical RSA signature bypass in OCaml's jose package leads this digest, alongside a Forgejo RCE, Windows Server RDS breakage, and a MikroTik KEV entry.
In this digest
Start with the free Posture Self-Check to see where you stand against the current threat landscape.
Free Posture Self-Check