Twitxy Saved

Nick VanGilder

Verified account This account is verified. Learn more

@nickvangilder

Director of Offensive Security Operations at Fortune 500 Bank
Nashville, TN linkedin.com/in/nickvangild… Joined March 2010
3.4K Following 4.3K Followers 2.3K Posts

Latest posts 5

Nick VanGilder Verified account @nickvangilder
I get the desire to support vulnerability researchers in the security community who are finding and publicly disclosing vulnerabilities, but at some point this hamster wheel is just not sustainable. Repeatedly manufacturing urgency around vulnerabilities _before their actual risk is understood_ is highly problematic IMHO. Beyond the loss of hard earned credibility when we do this, it’s also super taxing on security teams that often have to drop other important work because leadership is worried that the sky is falling and needs certain assurances.

A vulnerability can be technically severe without representing an immediate, “drop everything” threat to every organization. We can and should celebrate and hype that part…. but when every vuln is marketed like it’s the next Log4Shell, I do think that distinction gets lost. This is madness.
5 replies 6 reposts 45 likes 2.2K views View on X
Nick VanGilder Verified account @nickvangilder
To add more color to my other post, I’m not saying companies _shouldn’t_ invest in AI vulnerability discovery and accelerating patching. In certain cases, exploitation of a vulnerability in, say, an external API could immediately give an adversary access to sensitive data. To me, that’s a high priority system that should already be identified in advance and require special attention (including prioritized patching).

Not all systems and exploitation scenarios are created equal though. In other intrusions, initial compromise is only one stage of a much larger problem set. And for an adversary to accomplish whatever their objectives are, they often have to perform additional actions that introduce operational risks that create opportunities for defenders to detect and disrupt them before getting to the Crown Jewels.

So, to me, investing in getting better at detecting and disrupting adversaries in these kind of situations is likely to deliver a higher ROI than investing disproportionately in finding and patching vulnerabilities faster in all systems broadly. Especially when those detection and response capabilities can help defend against multiple vulnerabilities and attack paths. At some point, you can only patch so fast. And if AI continues to accelerate vulnerability discovery and exploitation, eventually that won’t be fast enough.
Nick VanGilder Verified account @nickvangilder
The AI labs would much rather sell you an AI-powered vulnerability discovery tool and convince you the answer to that problem is (yet) another tool to help you get faster at patching instead of telling you to get better at detecting and responding to the exploitation of those same vulnerabilities using the tools you already invested heavily in. I guess one is easier to monetize.