OpenAI, Anthropic AI agents targeted real people and systems in cyber tests
2026-08-05T07:23:21Z•01ac57bb1f6713bb622bf94339af64966a8bd789615ed23ceae8c7dec426e70d
CVE-2026-18577AI agent securityAPT29Android malwareChainDropClickFixMicrosoft 365Midnight BlizzardN-able N-centralOpen VSXRATWi-Fi attacksXCSSETactive exploitationadversary-in-the-middleauthentication bypassdata breachdeveloper targetingdevice-code phishinginfostealermacOS malwarenpmpasskey hijackingphishing-as-a-servicesupply-chain attack
What happened
BleepingComputer security headlines report active and emerging threats including exploited authentication bypasses, supply-chain compromises, phishing-as-a-service, macOS and Android malware, passkey hijacking, targeted attacks against Microsoft 365 accounts, and large-scale data exposure. The most immediately actionable item is active exploitation of CVE-2026-18577 in N-able N-central, while ChainDrop’s compromise of more than 1,300 npm packages represents a significant software supply-chain risk.
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- bleepingcomputer
- Record identifier
- 01ac57bb1f6713bb622bf94339af64966a8bd789615ed23ceae8c7dec426e70d
- Enrichment time
- 2026-08-05T07:23:21Z
- 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.