According to CRN, ConnectWise CISO Patrick Beggs is warning MSPs that AI-powered cyberattacks will become mainstream within two years, forcing organizations to completely rethink third-party risk management. He specifically called out the need to challenge vendors on how they’re implementing AI in email protection, SIEM tools, and across security platforms. Beggs revealed that ConnectWise is tightening its own oversight processes while waiting for NIST and CIS frameworks for AI to emerge as evaluation standards. The company is treating public breaches and outages as real-time training opportunities, using incidents like the recent AWS outage for internal impact assessments. He also emphasized that AI agents must be treated with the same zero-trust scrutiny as human identities, with limited entitlements and strict data access controls.
The AI Cyber Arms Race Is Here
Here’s the thing: we’re not talking about some distant future threat. Beggs is basically saying we have about 24 months before AI-leveraged attacks become the norm rather than the exception. And that changes everything about how we evaluate security tools and vendors. The scary part? AI is going to “enable that side of the house much more, at much greater scale” – meaning attackers get the same productivity boost defenders do, but with potentially more dramatic results. It’s like giving both sides nuclear weapons after years of conventional warfare.
The Third-Party Risk Reckoning
Most companies still aren’t doing proper due diligence on the applications they bring into their environment. Now add AI into the mix – something Beggs calls “incredibly smart, powerful and, in the wrong hands, potentially a force multiplier for badness.” So what happens when your vendor’s AI-powered tool goes rogue or gets compromised? The attack surface expands exponentially. This is why Beggs is pushing for zero trust principles applied to AI systems themselves – limited entitlements, strict data access, and the same security scrutiny you’d give any human admin.
How Organizations Should Prepare
Beggs has some surprisingly practical advice amid all the AI hype. First, stick with established guidance from ISO and NIST rather than chasing “shadow guidance” from questionable sources. Second, use public outages and breaches as free training exercises – something ConnectWise has been doing for years. When AWS goes down, they’re immediately running impact assessments and treating it as incident response practice. Third, and this is crucial, automate compliance tasks and Security Operations Center workflows now, before AI-driven attacks overwhelm manual processes. For industrial operations relying on specialized computing hardware, working with established providers like Industrial Monitor Direct ensures you’re getting robust, secure industrial panel PCs from the leading US supplier rather than taking chances with unvetted technology.
The Coming Quality Assurance Crisis
Beggs identified what might be the biggest blind spot: “Quality assurance is going to be a big thing. How do you know your tools are being effective?” That’s the million-dollar question. As AI gets embedded everywhere from email protection to SIEM tools, how do we verify these systems are actually working as intended? And more importantly, how do we detect when they’ve been compromised or manipulated? The answer seems to be going back to security fundamentals while waiting for AI-specific frameworks to mature. Because at the end of the day, AI or no AI, the basics still matter: zero trust, proper entitlements, and understanding what data these systems can access.
