CRO, Check Point Software Technologies
Interview with Itai Greenberg, Check Point’s Strategy Chief
Check out this sharp interview with Check Point’s Itai Greenberg on how GenAI is reshaping cybersecurity, and why Hybrid Mesh Security is the path forward.
During the interview, Greenberg unpacks how attackers are weaponizing GenAI — and what businesses must do to fight back.
About Itai Greenberg
Itai Greenberg, Chief Strategy Officer (CSO) and former Vice President of Product Management at Check Point Software, has been discussing the impact of generative AI on cybersecurity and the company's product strategy. In a June 2025 interview, Greenberg stated that attackers are using AI to create phishing emails and websites at scale with little effort. He described Check Point's approach to AI from three angles: the attacker, protecting AI systems, and using AI for prevention. He said Check Point offers products to control employee use of public large language models (LLMs) and to secure organizations' own GenAI applications. Greenberg also advocated for a "hybrid mesh" security architecture, arguing that customers need a platform that provides prevention and simplicity rather than multiple products and alerts. In earlier appearances, Greenberg discussed Check Point's Secure Service Edge (SSE) and Secure Access Service Edge (SASE) offerings, emphasizing a hybrid approach that performs inspection on the device or browser to reduce latency and privacy concerns. He commented that funneling all internet traffic to cloud providers for SSL decryption would be costly and create cross-border privacy issues. Greenberg also highlighted Check Point's partnership with Microsoft Azure, stating that Microsoft selected Check Point as the number one security vendor for Azure. He promoted Check Point's Security Log Analytics Service, which uses Azure Data Explorer to aggregate logs from Check Point enforcement points. Additionally, Greenberg warned that organizations should prepare for quantum computing attacks, noting that hackers are already collecting encrypted data for future decryption.
Transcript (10 segments)
I
Itai Greenberg0:00
We're looking at AI from three different angles. The first angle is the attacker. So attackers are using a lot of AI and now more and more they're attacking systems that are AI-driven. So that's the attacker side and many understand what does it mean. It's easy for them to create mails for the attackers because of LLMs. It's easier for them to create phishing websites and they can do it in scale with very little effort. So this is the attacker side.
On the prevention side, there are two aspects. One is how do we protect AI systems, LLMs, GenAI systems. So I'm an organization and as an organization I'm either using public LLM or I'm building my own application based on GenAI. So we need checkpoint. We are offering customers today product for both scenarios. One is if your employees are using ChatGPT but so many others LLMs, some of them are good, some of them are less good, you want to control it. It's the shadow IT of GenAI. What data they're sharing when they're communicating with those LLMs? Code, customers, finance, acquisitions. So you want to control what data, which ones they use and what data they actually share outside. So that's number one. And we now have product that gives you full visibility for all the public LLMs. What you use, are they good or bad? Are they proven to, are they going to use your data to train their systems or not? What data is going outside and controlling it. So that's one aspect.
The other aspect is says okay I'm building my own AI application. How do I protect this one? And you need to protect the prompt and you need to protect AI will talk to AI. So how do you protect my own application? And here we are offering multiple products from protecting the prompt, protecting the APIs, protecting the things that runs in the cloud, the serverless elements that runs in the cloud. A lot of technologies and integrating our security inside Nvidia to protect customer that are building their own AI application whether in the cloud or in their data center on prem.
The third element is how we use AI to make our product more productive, more efficient. And here we're doing quite a bit. We introduce AI ops to reduce operation cost and to make things more easy. So we're looking at it as a pyramid. Okay. At the bottom of the pyramid, we are living in a hyperconnected world. So users working everywhere. Applications are in the cloud or in SaaS. You still have branch offices and headquarter and sometimes data centers. You now have LLMs. Each environment is complex. Putting them all connected together is hyper complex. Okay. So, we're dealing with hyperconnected world that create hyper complexity. That's the reality.
Now, customers needs to make a decisions. What will be the architectures they use to secure and to connect everything and there are more than one architecture. Okay you can think of one architecture that says everything sits on prem, no one can work from home, I can send all traffic from everyone's to my data center and I will do the security in the data center. I can send all the traffic to the cloud and do the security in a SASE. Every architecture that you choose have, if you go with an architecture that is very rigid, very strict, you'll find yourself having problems. So the right architecture is what we call the hybrid mesh. Hybrid mesh says we understand that you live in many many places and so you need everything to be connected to everything and you need the security everywhere. So this is the hybrid mesh.
Once you choose the architecture for you as the customers, now you need to choose which vendors and which platform will help you to roll out to actually implement this architecture. And this is where Checkpoint steps in. Checkpoint comes and tell you if you choose hybrid mesh we can give you the best platforms to manage it which is the infinity platform which is one place to manage all the policy, one place to manage all the security, one place to share intelligence between all the product and especially around the foundation of connectivity, secure connectivity. That's the story of checkpoint. And it's in the cloud, it's the SASE, it's inside the browser, it's on premise, it's in the branch, it's in the OT environment. Everything is connected. And we live all in all those places. We have security network security in the cloud, network security in the SASE, network security in the browser, all those places.
I think that I would list three of them. One is definitely the AI. AI is the number one threats. We will have more hackers, more bad guys. They will use better tools. They will be way more efficient. They will generate more sophisticated attacks. And it's all because of AI. So AI will make the hackers 10 times, 100 times, 1,000 times better. The hackers, no question. And it will impact all of us.
The second thing is the complexity. Organizations are dealing with a lot of technologies and every day and the speed of new technologies and the amount of the new technologies is overwhelming and the complexity is getting bigger and bigger.
And the third challenge is the shortage of strong technical smart people that can help them to deal with problem number one and problem number two. And so what they need, they need security that says it's simple. You don't need to walk. You don't need to do the, don't alert me about problems. Don't tell me you have a problem here or you might have a problem there. Fix the problem for me or avoid. I don't want to deal with problems. I want solutions. And that's what customers really need. They need to have a platform. They need to have prevention. They need to have security. They don't need to have many many products. They don't need to have a lot of alert and visibilities. They need to have prevention.
Quantum is yet another one. Definitely another one out of many. Customers should start thinking about quantum attacks now not when it becomes available. The reason is because hackers, especially nationwide hackers, are collecting data right now, encrypted data. They don't know what to do with it. But when quantum encryption will be available, they will encrypt all the data they collected. And so it's not about having the right solutions when the hack is ready to break encryption, is to encrypt the traffic now to be ready for the days of quantum decryption is ready. And real checkpoint we are offering in every one of our product right now the ability to decrypt traffic that is quantum protected.