Huntress: ConnectWise M365 attack likely driven by AI agent, three modes spotted in one day
cyb3rops · x · 2026-10-27
Huntress published follow-up analysis of the ConnectWise Cloud Backup credential abuse that hit Microsoft 365 tenants. Microsoft attributed the activity to Storm-0940 using ARTEX, an AI agent framework; Huntress telemetry is consistent but not independently verified.
Their working hypothesis: an AI agent driving scripted tooling with occasional human intervention. On October 15 they observed three modes of operation (see timeline image):
- Scripted enumeration: same Teams-app procedure tenant after tenant; backup credential walking 1,000 tenants within 2.5 hours; fixed six-resource probe at 4 seconds per tenant.
- Hands-on operator work: secrets added/removed and Global Admin granted/dropped in two tenants, then 15 hand-picked tenants worked one at a time.
- Ghost execution: the Teams app deleted itself, yet runs kept firing and getting rejected for 22 minutes.
Huntress is low-to-medium confidence that an LLM was involved, citing cadence, retry loops, and user-agent patterns — but these AI indicators will fade as tooling improves, so defenders should detect what activity does, not how it paces itself.
More from Safety
- Researcher Fears Describing Agent Attack Scenario Might 'Summon It Into Existence' — jd_pressman · 2026-10-27
- AI Agents Make Attack Escalation Cheap as a 'Pizza Topping,' Researcher Warns — jd_pressman · 2026-10-27
- Dev invokes Second Amendment to defend open-source cyber models, serves them at cost — mimi10v3 · 2026-10-27
- Former OpenAI researcher questions why AI faces no mandatory liability insurance — Miles_Brundage · 2026-10-27
- Researcher says Apple sat on his iOS vulnerability credit for 8 months — kingslyj · 2026-10-27
- Samsung TVs screenshot your screen every 500ms; company settles Texas lawsuit over viewing-data sales — nikola_mr64990 · 2026-10-27