How AI Agents Can Trigger Runaway Costs for Enterprises
2026-09-22T02:51:34Z•1b3dd7c4b4724485b2bea06522bc7bd2aa2b754aa971e593cc9010fabd5214b7
CVE-2026-76460CVE-2026-85706AI agentsAI securityAPTAndroid banking trojanChina-linked threat actorsCisco ISEGitLab vulnerability managementLLM securityLinuxMFA bypassMaaSMicrosoft 365North KoreaOAuth abuseRussia-linked threat actorsWindowsagent hijackingdata extortionidentity securitymalwaremisalignmentphishingprompt injectionransomwarerunaway costs
What happened
Dark Reading feed covering September 2026 cybersecurity developments, including AI-agent abuse and misalignment, runaway LLM consumption, browser-agent hijacking, AI-enabled fraud and attacks, OAuth consent abuse, ransomware and data extortion, malware campaigns, APT activity, and major vulnerabilities. Notable vulnerabilities include Cisco ISE authentication bypass CVE-2026-76460 and maximum-severity GitLab path traversal CVE-2026-85706; additional reporting describes actively exploited Microsoft flaws, Cisco vulnerability chaining by Sandworm, and emerging phishing, Android banking, and MaaS
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- darkreading
- Record identifier
- 1b3dd7c4b4724485b2bea06522bc7bd2aa2b754aa971e593cc9010fabd5214b7
- Enrichment time
- 2026-09-22T02:51:34Z
- 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.