🔍 Read the full analysis: Cyberattacks Are Moving Faster With AI: A Practical Response Guide on ThorstenMeyerAI.com
Get hardware and tech essentials delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
U.S. agencies say attackers have used AI-generated scripts against internet-exposed industrial controllers, and Microsoft reported that an AI agent helped Storm-0940 target cloud tenants through a compromised backup app. Microsoft said most targeting failed, and it did not establish that every successful sign-in led to data access. The reports point to faster attack activity, while recommended defenses remain inventory, patching, stronger sign-in controls and practiced response plans.
U.S. agencies and Microsoft have described AI-assisted cyber activity targeting industrial control systems and cloud organizations, including an operation that reached privileged access in some cases in under three hours. The accounts show software speeding up familiar attack steps, but Microsoft said most of the targeting it tracked failed and cautioned that a successful sign-in does not by itself prove data was accessed.
A joint advisory published August 18 by the NSA, CISA, FBI, Department of Energy and Environmental Protection Agency says actors used AI-generated exploitation scripts against internet-reachable Siemens S7 programmable logic controllers. The devices, known as PLCs, help operate industrial processes. The advisory says the scripts were disguised as monitoring tools and built using the open-source Snap7 library. Agencies assess the activity as likely preparation for operational effects against critical infrastructure; that assessment is not confirmation that physical disruption occurred.
In an October 21 report, Microsoft Threat Intelligence said the China-state-aligned actor Storm-0940 used ARTEX, an open-source agentic penetration-testing framework, in activity connected to the compromise of ConnectWise Cloud Backup. Microsoft said the actor used credentials associated with the backup application to attempt sign-ins to more than 2,250 tenants. Its report also recorded over 151,000 failed logons from October 10 to 14, with a peak of 2,900 failures per minute; Microsoft said none of those attempts succeeded.
Microsoft said some attacks moved from initial targeting to privileged cloud access in less than three hours. It also stressed that most targeting was unsuccessful and that an affected application or successful sign-in does not necessarily mean data was accessed. ConnectWise had disclosed the compromise on October 16. The company said existing backup archives showed no indication of compromise and that it suspended Cloud Backup, SaaS Security and Cloud Management while investigating.
Infographic 01 · The record so far
AI is in the attacker’s toolchain. On the record.
Six documented events in twelve weeks. Each label says what the source itself states about AI. Figures are the vendors’ and agencies’ own.
- 30 Jul 2026Unit 42: AI on attackDeepSeek inside the Hermes Agent framework picks and switches targets on its own. Limited impact: across autonomous and manual attempts, data theft from 3 NetScaler systems confirmed. Single-vendor report.
- 18 Aug 2026NSA, CISA, FBI, DOE, EPA: AI on attackAI-generated exploitation scripts, disguised as monitoring tools, against Siemens S7 PLCs in US critical infrastructure. “Not a theoretical risk.”
- 10–14 OctStorm-0940 password spray: AI on attack151,000+ failed logons across eight organizations. No successful sign-in. Lockouts made up nearly 98% of responses.
- 16–21 OctConnectWise Cloud Backup: AI on attackStolen app credentials, then machine-speed sign-ins to customer tenants. Microsoft attributes it to the China-state aligned Storm-0940.
- 21 Oct 2026OpenAI, Patch the Planet: AI on defenceAI found the critical PHP flaws before attackers did. OpenAI expects similar findings to “increase sharply in the coming months.”
- 21–22 OctQuest Diagnostics: no AI link reportedUnauthorized access to a lab ordering and results system; certain patient records accessed. No cause or attacker named.
Most AI-driven attempts in these reports failed. What changed is the cost of trying.Assessment: Thorsten Meyer
Sources: Unit 42 (30 Jul 2026); NSA, CISA, FBI, DOE, EPA advisory (18 Aug 2026); Microsoft Threat Intelligence (21 Oct 2026); ConnectWise Trust Center (from 16 Oct 2026); OpenAI, Patch the Planet (21 Oct 2026); Quest Diagnostics Form 8-K (22 Oct 2026).
Powered by Thorsten Meyer AI
Faster Attacks Raise Response Pressure
The reports matter because they describe AI as a way to accelerate reconnaissance, scripting and repeated login attempts, rather than as evidence of a new kind of attacker that can defeat every defense. Automation can let an operation run longer and perform routine steps more quickly. For defenders, that can reduce the time available to notice activity, contain an account or block access to a connected service.
The potential consequences differ by target. An exposed industrial controller can affect systems that support manufacturing, energy, water and other infrastructure. A compromised trusted cloud application can give an attacker a route to multiple customer environments. The source reports describe risks and activity, not confirmed physical damage or proven data theft across all affected tenants.
That distinction also shapes the practical response. Agencies and security teams are not advising organizations to rely on AI tools alone. Their guidance emphasizes asset inventories, timely patches, restricted remote access and tested incident plans. Those measures can help limit the reach of both automated and human-led attacks.
hardware security keys FIDO2 for Microsoft 365 MFA
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Two Reports, Different Targets
The August advisory concerns Siemens S7 PLCs, including models from the S7-200 through S7-1500 and safety controllers. Agencies say actors found exposed devices using internet-scanning services. Their recommendations include inventorying controllers, updating firmware, keeping PLCs off the public internet, blocking TCP port 102 at the perimeter, and separating operational technology from corporate IT. They also advise multi-factor authentication for remote OT access and monitoring S7 traffic for writes outside approved change windows.
Microsoft’s October account concerns a trusted software supply chain. ConnectWise first disclosed its Cloud Backup compromise on October 16. Microsoft later attributed the activity to Storm-0940 and described ARTEX as a multi-agent framework capable of carrying out stages from reconnaissance to data exfiltration with limited human involvement. Attribution and technical assessments in the report are Microsoft’s conclusions.
A separate July 30 report from Palo Alto Networks’ Unit 42 described an actor using the DeepSeek model within the Hermes Agent framework, controlled through Telegram. Unit 42 said the agent performed extensive targeting analysis quickly. These examples indicate interest in applying AI to existing attack methods; they do not establish that every AI-assisted operation succeeds. OpenAI, meanwhile, said AI had identified critical PHP vulnerabilities that were patched on October 21 and forecast that AI-found critical flaws would increase sharply in the coming months.
“The majority of the targeting wasn’t successful.”
— Microsoft Threat Intelligence
business firewall with intrusion prevention system
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Confirmed Access, Unsettled Impact
The available reports do not establish that the industrial-controller activity caused operational disruption or physical damage. The agencies describe likely preparation for operational effects, an assessment of intent and capability rather than confirmation that effects occurred.
For the ConnectWise-related activity, Microsoft said some attacks reached privileged cloud access, but a successful sign-in does not necessarily show that data was accessed. ConnectWise said its existing backup archives showed no indication of compromise. The full scope of any customer impact, including whether information was obtained from particular tenants, remains unclear in the supplied accounts.
Forecasts also need to be separated from observed activity. OpenAI expects AI-found critical vulnerabilities to rise sharply, while a security leader quoted in the source material predicted a near-term “vulnerability tsunami.” That leader’s view is an expert opinion, not a measured forecast; the precise scale and timing of any increase are not established.
managed network switch VLAN for industrial control systems
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Patch, Review and Test Defenses
Organizations should first check whether internet-facing systems, cloud applications or third-party integrations create routes into sensitive environments. The agencies’ controller guidance calls for a complete PLC inventory, firmware updates, blocked public access and separated IT and OT networks. Remote OT access should use multi-factor authentication, and teams should watch for unexpected controller commands.
Cloud and identity teams can review third-party application permissions, remove access that is no longer needed, and investigate unusual sign-in activity. Boards and executives should ask for clear patch deadlines, phishing-resistant sign-in where available, and incident-response exercises that test how quickly teams can disable credentials or isolate affected systems. These are baseline measures, not a guarantee against compromise.
For individuals, the source recommends starting with the email account: use a unique password, enable strong multi-factor authentication and review recovery methods and connected applications. Further details about the ConnectWise investigation and any confirmed customer impact would clarify the scope. Until then, organizations should treat the reports as evidence of faster attack activity, while avoiding assumptions that every attempted or successful login resulted in data loss.
encrypted external hard drive for offline backups
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
Did the reported AI-assisted attacks cause confirmed physical damage?
The supplied reports do not confirm physical damage or operational disruption. U.S. agencies said activity against industrial controllers was likely preparation for operational effects.
Did Storm-0940 access data from all 2,250 tenants?
No. Microsoft reported attempts to sign in to more than 2,250 tenants, but said most targeting was unsuccessful. It also cautioned that a successful sign-in does not necessarily mean data was accessed.
What should organizations do first?
Start with an inventory of internet-facing and sensitive systems, apply available patches, restrict remote access, review third-party application permissions and test incident-response procedures.
Is a sharp rise in AI-found vulnerabilities confirmed?
No. OpenAI said it expects AI-found critical vulnerabilities to increase sharply in the coming months. That is a forecast, not a documented measurement of a future increase.
Source: ThorstenMeyerAI.com
Evergreen bestsellers Picks
bestsellers
As an affiliate, we earn on qualifying purchases.
