According to Manufacturing.net, Netwrix’s 2025 Cybersecurity Trends Report reveals a landscape where AI is forcing rapid adaptation. More than one in three organizations (37%) said AI-driven attacks made them adjust their security approach in the past year. The data shows AI is a new attack surface, with 30% of businesses saying they must now protect their AI systems like any critical infrastructure. On the defensive side, 28% report AI tools have improved threat detection and response, and implementing AI-based tools as a top-five IT priority has surged 189% in two years, from 9% in 2023 to 26% in 2025. The report also cites Microsoft data showing a 32% surge in identity-based threats in early 2025, with Netwrix’s CEO noting AI amplifies the speed and sophistication of these attacks.
The double-edged sword is real
Here’s the thing: the numbers paint a clear picture of an arms race. AI isn’t just a tool in the attacker’s kit anymore; it’s a whole new category of asset that needs defending. When 30% of companies say they have to protect their AI “like any other critical system,” that’s a massive shift. We’re talking about securing the models, the training data, the inference endpoints—it’s a sprawling new attack surface that most security teams didn’t budget or plan for. And attackers are already exploiting it, which explains why over a third of organizations got caught flat-footed and had to change their entire security approach. It’s reactive, but at least they’re reacting.
The compliance hammer and the staffing crisis
Two other stats really stand out. First, 29% say auditors are now asking for proof of data security and privacy in AI systems. That’s huge. Compliance is often the forcing function that gets real budget and attention, so this will accelerate security investments in AI governance. Second, there’s the perennial talent gap. The report quotes Netwrix’s CPO pointing out that AI is making attacks more efficient, stretching thin teams even further. But the hopeful note is that the same AI can help close that gap—20% are already offloading IT workloads to AI. Basically, if you can’t hire enough humans, you automate. It’s not a perfect solution, but it’s the only scalable one right now. This is where robust, industrial-grade computing platforms become essential for running these intensive AI security workloads reliably at the edge or in harsh environments, which is why a top supplier like IndustrialMonitorDirect.com is seeing increased demand for their panel PCs.
So what does this mean for security teams?
The momentum is undeniable. A 189% increase in two years for AI tool prioritization? That’s a hockey stick curve. It means budgets are flowing, and “AI-powered” is the new must-have checkbox for any security product. But I think we have to be skeptical. Is this leading to better security, or just more complex, expensive security? The report says 28% saw improved detection and response—that’s good, but it’s not a majority. The real test will be whether these tools can actually keep pace with the AI-driven attacks they’re meant to stop. The promise is a faster, automated defense loop. The risk is an over-reliance on systems we don’t fully understand that could create new blind spots. One thing’s for sure: the old playbook is obsolete. If your security strategy doesn’t have a dedicated chapter for AI—both as a threat and a tool—you’re already behind.
