Responding To Cyberattacks In An AI-Accelerated World

🔍 Read the full analysis: Responding To Cyberattacks In An AI-Accelerated World on ThorstenMeyerAI.com

Before you orderOffer from Amazon

Get the latest gadgets delivered free with Prime

• Fast, free delivery on millions of items
• Prime Video, Amazon Music and more included
• Member-only deals all year

Start your free Prime trial →Free trial for eligible customers · Cancel anytime

As an affiliate, we earn on qualifying purchases.Hide

US agencies and Microsoft have reported AI-assisted cyber activity targeting industrial control systems and cloud accounts, including attackers’ use of generated scripts against Siemens controllers and an agent framework in a supply-chain-related incident. The reports document new uses of automation, but they do not show that every targeted system was compromised or that data was accessed in every affected cloud tenant.

A joint advisory issued on August 18 by the NSA, CISA, FBI, Department of Energy and Environmental Protection Agency says threat actors used AI-generated scripts disguised as monitoring tools against internet-reachable Siemens S7 programmable logic controllers. The agencies describe reconnaissance and capability development involving models from the S7-200 through S7-1500, including safety controllers, and assess that the activity may be preparation for operational effects against critical infrastructure.

Microsoft Threat Intelligence reported on October 21 that Storm-0940, which it describes as a China-state-aligned actor active since 2021, used ARTEX, an open-source multi-agent penetration-testing framework. Microsoft linked the activity to the compromise of ConnectWise Cloud Backup and said the framework was used to attempt sign-ins across more than 2,250 tenants after credentials for the cloud backup application were obtained. From October 10 to 14, Microsoft recorded more than 151,000 failed logons, with a peak of 2,900 per minute; none of those attempts succeeded.

Microsoft said some attacks reached privileged cloud access in under three hours, while stressing that most targeting was unsuccessful. It also cautioned that an affected app or successful sign-in does not necessarily mean data was accessed. ConnectWise said existing backup archives showed no indication of compromise and said it suspended Cloud Backup, SaaS Security and Cloud Management while investigating.

Faster Operations Raise Response Pressure

The reports point to a change in the speed and scale of familiar attack steps, not proof of an autonomous system capable of reliably breaking into any target. Scanning, password spraying and testing public exploits remain recognizable methods; automation can let operators repeat or adapt them across more targets and at greater speed.

That matters to organizations because defensive work still depends on knowing what is connected, closing known weaknesses and limiting access. An attack that moves from initial targeting to privileged cloud access in hours can leave less time for investigation and containment. In operational technology, an exposed controller can affect systems that support production or public services, so network separation and controlled remote access matter alongside software fixes.

AI can also help defenders identify flaws. OpenAI said it found critical vulnerabilities in PHP that were patched on October 21, and forecast that AI-found critical vulnerabilities will “increase sharply in the coming months.” That is the company’s expectation, not a measured future trend. If discoveries rise, organizations may face a larger and faster-moving queue of patches; the source material does not establish how many vulnerabilities will be found or how quickly attackers might exploit them.

Three Reports Track AI Use

The August advisory focuses on industrial control systems. It says actors scan for exposed Siemens PLCs and use scripts built with the open-source snap7 library. The agencies recommend inventorying controllers, applying firmware updates, keeping PLCs off the public internet, separating IT and operational technology networks, requiring multi-factor authentication for remote OT access and monitoring S7 traffic for writes outside approved change windows.

Microsoft’s October account concerns a different route: a trusted cloud application and its credentials. The company says ARTEX can support activity from reconnaissance through data exfiltration with limited human interaction. That description is Microsoft’s characterization of the framework; it does not establish that all stages occurred in every incident. The ConnectWise compromise was disclosed by the company on October 16, before Microsoft published its report.

A separate July 30 report from Palo Alto Networks’ Unit 42 described a Chinese-speaking actor using the DeepSeek model with the Hermes Agent framework, directed through Telegram. Unit 42 said the agent conducted extensive targeting analysis in minutes. Together, the reports describe different tools and incidents; they do not establish that all activity was carried out by the same actor or through a single coordinated campaign.

“This is not a theoretical risk—it is an active threat.”

— Joint advisory from the NSA, CISA, FBI, DOE and EPA

Access Does Not Prove Data Theft

The available reports leave important questions open. Microsoft says a successful sign-in or affected application does not necessarily indicate data access; the full scope of any access and whether information was taken are not established here. ConnectWise said its existing backup archives showed no indication of compromise, but the source material does not provide a final investigation finding or a complete account of customer impact.

The government advisory assesses that activity against PLCs is likely preparation for operational effects. That is an assessment of intent, not confirmation that the attackers caused physical disruption. The source material also includes a security leader’s prediction of a “vulnerability tsunami” within months and an estimate of roughly 18 months to prepare. Those figures are expert opinion, not measured evidence, and the identity and supporting methodology are not supplied.

Details about the full number of affected organizations, the results of the ConnectWise investigation, and whether the reported industrial activity led to operational consequences remain unclear. The reports also do not establish that AI was necessary for the attacks, only that tools using AI were involved in the described activity.

Investigations and Basic Controls

ConnectWise’s investigation will help clarify whether its suspended services or customer environments suffered data access beyond the information disclosed so far. Organizations using affected services should follow vendor updates, review authentication and application activity, and assess whether third-party access remains necessary. Microsoft’s report does not establish that every tenant targeted through the app was breached.

For industrial operators, the agencies’ recommendations offer immediate checks: identify every PLC, patch firmware, remove public internet exposure, segment operational networks and protect remote access with multi-factor authentication. Companies should also review third-party applications and rehearse incident response, rather than treating AI adoption as a substitute for basic security work.

For individuals, securing the email account is a practical starting point because it can be used to reset access to other services. Use a unique password and phishing-resistant sign-in where available, and review connected apps. The next confirmed milestones are further investigation findings and any new agency or vendor advisories; the timing and scale of the predicted rise in AI-discovered vulnerabilities remain uncertain.

Key Questions

Did the reports confirm that attackers stole data?

No. Microsoft said an affected application or successful sign-in does not necessarily mean data was accessed. ConnectWise said existing backup archives showed no indication of compromise; the full investigation outcome is not included in the available material.

What did US agencies report about industrial systems?

A joint advisory said actors used AI-generated scripts against internet-reachable Siemens S7 programmable logic controllers. The agencies described the activity as an active threat and assessed it as likely preparation for operational effects, not confirmation that disruption occurred.

What is ARTEX?

ARTEX is an open-source multi-agent penetration-testing framework. Microsoft said Storm-0940 used it in activity involving ConnectWise Cloud Backup and cloud tenant sign-in attempts.

Is a surge in AI-found vulnerabilities confirmed?

No. OpenAI forecast that AI-found critical vulnerabilities will increase sharply in coming months after saying it found PHP flaws patched on October 21. The future increase is a forecast, not an established measurement.

Source: ThorstenMeyerAI.com