AI models are now finding vulnerabilities faster than teams can patch them. When hundreds or thousands of new flaws surface, patch cycles fall behind and exposed systems stay open until a fix ships. The question becomes what stops a flaw from being executed in the meantime.
In this interview on TFiR, Steve Winterfeld, Advisory CISO at Akamai, breaks down how to put mitigations and segmentation around vulnerabilities that cannot be patched fast enough, and how Akamai’s privacy officer thinks about trust in AI.
Guest: Steve Winterfeld, Advisory CISO at Akamai
Show: CISO Insights
LinkedIn: Steve Winterfeld
Watch the discussion on YouTube
Here is what every security leader needs to know.
Technical Deep Dive
Q: How do you protect systems when vulnerabilities are found faster than you can patch them?
Steve Winterfeld, Advisory CISO at Akamai, says frontier models are discovering hundreds or thousands of new vulnerabilities, and teams will not be able to keep up by patching alone. Winterfeld’s answer is to put protection around patching, so a vulnerability is still discovered or mitigated before it can be executed.
“If we’re discovering hundreds or thousands or whatever the number is of new vulnerabilities, we’re not going to be able to keep up with that. And so you need that protection around patching.”Steve Winterfeld, Advisory CISO, Akamai
Q: What mitigations stop a vulnerability from being executed before it is patched?
Winterfeld frames the question as what mitigations are in place so that a vulnerability is quickly discovered or mitigated before execution. He points to classic web application firewalls, AI firewalls and many other applications as the places where that protection sits.
“what mitigations do you have that if you have a vulnerability, it’s still going to be either quickly discovered or mitigated before it can be executed.”Steve Winterfeld, Advisory CISO, Akamai
Q: How does internal segmentation limit the impact of a vulnerability that gets through?
Beyond perimeter mitigations, Winterfeld asks what is being done internally to segment systems. The goal is to minimize the impact when a vulnerability does get through.
“And then internally, what are you doing to segment those to minimize when it does get through what that impact is?”Steve Winterfeld, Advisory CISO, Akamai
Q: What did Akamai learn from the Mythos project Glasswing?
Akamai was part of the Mythos project Glasswing, which looks at how frontier models are doing. Winterfeld says the report shares Akamai’s lessons learned from that work, including the point that vulnerability discovery now outpaces patching.
“Akamai was part of the Mythos project Glasswing and how frontier models are doing. And so we talk a little about our lessons learned there.”Steve Winterfeld, Advisory CISO, Akamai
Q: How should organizations approach AI privacy and trust beyond GDPR and state laws?
Winterfeld notes that an Akamai chief technologist wrote about MCPs, while Akamai’s privacy officer wrote about how to think about the challenge of AI. The focus is not the specifics of executing privacy but what is trusted, how it is trusted, and which questions to ask to manage the program.
“Instead of saying follow GDPR and follow this California law, it’s really step back and think about how are we going to approach this in the long run?”Steve Winterfeld, Advisory CISO, Akamai
Resources & Documentation
- Account Protector, Akamai’s account protection product
- Bot Manager, Akamai’s bot management product
- API Security, Akamai’s API security product
***
👇 Click to Read Full Raw Transcript
Swapnil Bhartiya: You know, you folks of course do a lot of work in this space. Talk a bit about what else is there which may align with the findings of this report or the improvement that you see in the project that Akamai is internally working on to also help customers secure their posture.
Steve Winterfeld: You know, a couple areas that we touch on. One is Akamai was part of the Mythos project Glasswing and how frontier models are doing. And so we talk a little about our lessons learned there. And, you know, if we’re discovering hundreds or thousands or whatever the number is of new vulnerabilities, we’re not going to be able to keep up with that. And so you kind of need that protection around patching. If, you know, we’re not able to patch at the same speed vulnerabilities are being found. And so as we think about how to do that patching, you know, what mitigations do you have That if you have a vulnerability, it’s still going to be either quickly discovered or mitigated before it can be executed. And that’s on your classic web application firewalls. It’s on your AI firewalls. It’s on so many different applications. And then internally, what are you doing to segment those to mitigate it before it can be executed? And that’s on your classic web application firewalls. It’s on your AI firewalls. It’s on so many different applications. And then internally, what are you doing to segment those to minimize when it does get through what that impact is? Well, we talk a little bit about some of the lessons learned Throughout that. Another one that I found fascinating is our, you know, our— one of our chief technologists was the one that wrote about MCPs, but we also had our privacy officer write in, and not so much about, you know, the specifics of how to execute privacy, but how to think about, you know, the challenge of AI, how to think about You know, what is trusted? How is it trusted? What are the right questions to ask yourself to manage this program? And I think that’s also a good kind of an academic thought leadership way to go about it. Instead of saying, you know, follow GDPR and follow, you know, this California law, it’s really, you know, step back and think about How are we going to approach this in the long run?





