
BeyondTrust suspects Armory Castle uses AI to model target products for exploit development
Armory Castle began exploiting BeyondTrust zero-day CVE-2026-102764 on August 29, 2026.
Our coverage: 69 reports since October 8, 2026. Latest: BeyondTrust suspects Armory Castle uses AI to model target products for exploit development, October 18, 2026.

Armory Castle began exploiting BeyondTrust zero-day CVE-2026-102764 on August 29, 2026.

The rule, named autocad_orphaned_vba_project, detects orphaned VBA projects hidden inside DWG files.

Tech Times notes that iMazing’s scanner, Jamf Pro MDM audits, and the RosettaCheck database can identify Intel‑dependent apps, and 9To5Mac advises auditing fleets now before upgrading.

Combining the authentication bypass and command‑injection flaws lets attackers on the same network gain admin access and run commands.

SentinelOne, app.opencve.io, PremierePC Technology Group and TheHackerWire report that the flaw carries a CVSS v3.1 base score of 8.8, rated as high severity.

The flaw carries a CVSS score of 8.8 and is rated Important by Microsoft.

Palo Alto Networks’ Unit 42 finds threat actors using AI to slash breach cycles to under an hour, while its service integrates OpenAI and Anthropic models to detect vulnerabilities continuously

Armory Castle, tracked by Mandiant and Google Threat Intelligence, used the flaw starting August 29 to deploy backdoors and harvest credentials, with patches now available.

Anthropic assigned the vulnerability a CVSS v4.0 score of 8.7, equivalent to host code execution, according to CSO Online and noma.security

Patched releases include Drupal core 10.6.19, 11.3.19, 11.4.10 and Entity List module 3.0.10.

The California Department of Education disclosed the breach on October 16, 2026.

Google warns of continued targeting of edge devices as the group deploys webshells, backdoors and SOCKS5 tunneling after patch release