A Practical Guide To AI And The Faster Pace Of Cyberattacks

🔍 Read the full analysis: A Practical Guide To AI And The Faster Pace Of Cyberattacks on ThorstenMeyerAI.com

Buying for a business?Offer from Amazon

Get business pricing on tech for your team

• Business-only prices and quantity discounts
• Tax-exempt purchasing
• Multiple users, one account, clear invoices

Create a free Business account →Free to register

As an affiliate, we earn on qualifying purchases.Hide

US agencies and Microsoft have documented AI-assisted cyber activity involving industrial controllers and cloud services, adding evidence that attackers are using AI tools to automate parts of existing attack methods. The reports describe attempts against critical infrastructure and hundreds of organizations, but also note that most targeting in Microsoft’s account was unsuccessful and 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 threat actors have used AI-generated scripts against internet-reachable Siemens S7 programmable logic controllers, or PLCs. These industrial computers can control equipment and processes. The advisory says attackers used scripts disguised as monitoring tools and built with the open-source snap7 library. The agencies describe the activity as active and assess it as likely preparation for operational effects, not proof that such effects occurred.

Microsoft Threat Intelligence reported on October 21 that Storm-0940, which Microsoft describes as aligned with the Chinese state, used the open-source ARTEX agentic penetration-testing framework in activity associated with the ConnectWise Cloud Backup compromise. Microsoft says the actor attempted to access more than 2,250 tenants using credentials for the cloud backup app. It also recorded more than 151,000 failed sign-ins during a password-spraying campaign from October 10 to 14, with a peak of 2,900 failures per minute. Microsoft says none of those attempts succeeded.

Microsoft reported that some attacks reached privileged cloud access in under three hours, while stressing that most targeting was unsuccessful. It also cautioned that an affected application 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 during its investigation.

Faster Attacks Raise Defense Pressure

The reports point to a practical change in the speed and scale of routine attack steps, rather than evidence of an entirely new kind of attacker. AI tools can help automate reconnaissance, generate or adapt scripts, and run repeated sign-in attempts. That can shorten the time defenders have to detect activity and respond, particularly when an attacker abuses a trusted third-party application or reaches systems that are exposed to the internet.

For industrial operators, a compromised PLC could affect physical processes, although the advisory does not say that the reported activity caused operational disruption. For cloud customers, third-party app credentials can create a route into multiple tenants. The ConnectWise case shows why organizations need to review what access vendors and applications hold, rather than treating a trusted connection as risk-free.

AI can also aid defenders. The source material says OpenAI reported that AI identified critical PHP vulnerabilities that were patched on October 21 before attackers exploited them, and the company expects similar findings to increase sharply in the coming months. That is a company forecast, not a count of future vulnerabilities. More discoveries could improve security while also adding work for teams that must verify, prioritize and install fixes.

Three Reports, Different Evidence

The August advisory concerns industrial control systems, while Microsoft’s October report concerns a cloud service and its customer tenants. They are separate cases, with different targets and evidence. The advisory says actors found exposed Siemens controllers through internet scanning services and used AI-generated scripts. It names PLC models from the S7-200 through the S7-1500, including safety controllers, and identifies manufacturing, energy, water, chemical, food and agriculture, and commercial facilities as sectors of concern.

Microsoft’s account links Storm-0940’s activity to a supply-chain compromise involving ConnectWise Cloud Backup. The framework Microsoft describes can run multiple stages, from reconnaissance toward data exfiltration, with limited human interaction. The reported tenant targeting and failed logons show the scale of activity; they do not establish that every targeted organization was breached or that data was taken.

The source material also cites a July 30 Unit 42 report about a Chinese-speaking actor using the DeepSeek model with the Hermes Agent framework and Telegram. The available account says the agent searched GitHub after an exploit attempt failed, but provides no further outcome here. OpenAI’s comments about future vulnerability discovery and a security leader’s forecast of a coming “vulnerability tsunami” are projections, not measured attack statistics.

“The majority of the targeting wasn’t successful.”

— Microsoft Threat Intelligence

Impact and Forecasts Remain Unsettled

The available reports do not establish that the Siemens PLC activity caused physical disruption, or that the cloud activity resulted in data theft. Microsoft says an affected app or successful sign-in does not necessarily indicate that data was accessed. ConnectWise said its existing backup archives showed no indication of compromise, but the source material does not provide a final investigation finding covering every potentially affected system or tenant.

It is also unclear how much of the activity was directly generated by AI rather than supported by AI tools at particular stages. The source material does not give a consistent measure of how much faster these operations were than comparable non-AI activity. Predictions of a near-term surge in vulnerabilities or attacks should be read as forecasts; the stated “18 months” and “vulnerability tsunami” are an individual security leader’s opinion, not a verified timeline or statistical finding.

Prioritize Access and Patch Controls

Organizations should follow the controls in the government advisory and review their own exposure. The agencies recommend inventorying PLCs, patching firmware, keeping controllers off the public internet, blocking TCP port 102 at the perimeter, and separating IT and operational technology networks. They also call for multi-factor authentication on remote OT access and monitoring S7 traffic for writes outside approved change windows.

For cloud environments, the immediate work is to review third-party app permissions and credentials, check sign-in and audit logs, and confirm that incident-response teams know how to revoke access quickly. Across both settings, companies should maintain an accurate asset inventory, set patch deadlines, use phishing-resistant sign-in where possible, and practise incident response. Individuals can start by securing their email accounts, which often serve as a recovery route for other services.

Further clarity depends on the outcomes of investigations and any additional advisories or threat reports. Organizations should distinguish confirmed incidents from attempted access, and track future vulnerability findings without treating forecasts as established counts.

Key Questions

Have AI-driven attacks caused confirmed physical damage to industrial systems?

The cited joint advisory reports AI-generated scripts used against Siemens PLCs and assesses the activity as likely preparation for operational effects. The source material does not confirm physical damage or operational disruption.

Did the ConnectWise activity expose customer data?

The reports summarized here do not establish that customer data was accessed. Microsoft says a successful sign-in does not necessarily mean data was reached, and ConnectWise said existing backup archives showed no indication of compromise.

What does ARTEX do in Microsoft’s account?

Microsoft describes ARTEX as an open-source, multi-agent penetration-testing framework capable of supporting activity from reconnaissance toward data exfiltration with limited human interaction. Its use does not by itself show that every stage succeeded.

What should organizations do first?

Start with known assets and access: inventory internet-facing systems and industrial controllers, patch firmware, restrict remote access, review third-party app permissions, and check sign-in logs. The government advisory also recommends network separation and monitoring of industrial control traffic.

Is a surge 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; the source material does not provide a measured future count.

Source: ThorstenMeyerAI.com