Speaker 1: Let's get the latest on the hack now, which is giving cybersecurity stocks a bump today. My next guest warned about this when he was on the show two weeks ago, since data protection is crucial now because data leaks are used to train these AI models in China and elsewhere. Ivan Cerini is CEO of Farood, a web security monitoring platform. Ivan, let's start with what we know about DeepSeek's hack today. Look, I'm hearing reports they're restricting signups only to mobile users in China. What's going on?

Speaker 2: Yeah, thanks for asking. That's a great question. I would say, first of all, we need to wait until the actual information gets public about what actually happened. Was it a hack? Was it a breach? Or was it a massive surge of users on DeepSeek's website? We don't know yet, but definitely DeepSeek is disrupting the AI industry overall. It has been compared, obviously, to the Sputnik moment, and it could be a hack or it could be just a significant load of new users that they're acquiring. So let's wait for actual information from them.

Speaker 1: Sure. I was thinking of you this morning when I was toying with downloading the app. For the past week, I've thought to myself, do I need to get TikTok off the phone? So remind me about the methods that you described that TikTok is using to collect our activity on the phone and whether you think downloading DeepSeek, which is now number one on the App Store, would expose U.S. users to the same problem.

Speaker 2: So something that the guests just were mentioning that is really important to your question on the TikTok and how it collects information. So what is really important to remember that AI does not live on its own. It has four main pillars, which is energy, processing power, the chips that you discussed, and also the ingenuity or the engineering part that humans do. And also really important is the information and the data that AI uses to train itself, but also to become really relevant. And that's what is really important. What you said is, hey, how does TikTok or any other company that is involved or could be collecting information that can be used to train AI and how do they get that information? So obviously from the apps like TikTok or other apps could be or are potentially collecting information that can be used to train AI. But also really is important that really is not being appreciated as much yet is the value of the English-speaking information that all kinds of apps, whether they're collecting it or if they're connected to China or not, or pixels trackers, that's what our specialty is, what they do. They collect information that is very, very current. Like, Kelly, something you were doing perhaps before coming into the office today. When you're looking at something on a website, they are seeing and they can be using that to fine-tune recommendations or fine-tune or potentially weaponize it. So there's definitely a cybersecurity threat in information that is being collected and exported to China.

Speaker 1: And I think a lot of this hangs on the word can versus the word is. Do all of these Chinese apps just have the capability to do that or are they definitely doing that?

Speaker 2: Well, that's probably a trillion-dollar question. There is really – it's impossible to pinpoint and prove anything until it is proven to be true. And, for example, when it comes to data or information that is being collected by companies associated with China, can they be transferred to Chinese government? According to Chinese law, yes, it can be. So that's very important. When that information is used for training AI models, that collection of that information can potentially disrespect all of the laws we have in America or in other countries, and that can give a significant advantage to the Chinese innovation in AI.

Speaker 1: Right, because English language activity is highly valuable because English, at least for now, is still the forefront kind of leading edge of these AI models. So, as I mentioned, a lot of people have downloaded this app onto their phones. Ivan, should they delete it?

Speaker 2: Well, it's a personal choice. We live in a free world, so it's really up to them, to individuals, to make a choice. I certainly do not use the app myself because I don't want to be exposing my information. That's my personal opinion. With that said, there's also an interesting comment that I want to add. Just like in financial markets, you don't want to be making investment decisions based on outdated data, whether it's six months or a year or a couple of years old, because that wouldn't be a wise investment decision. Same with AI. When AI uses outdated data and outdated information, it's not going to be as accurate as if it is using the data that is very, very current. So, that's a really important value of the current information, potentially even protecting or preventing export of current information of Americans to adversarial countries. That's super interesting.

Speaker 1: So, it sort of suggests, like, if you don't want them to get the edge right now, maybe take this stuff off your phone right now. I don't know. I'll have to think about it. Ivan, thanks for rejoining us. Good to see you again. We really appreciate it today.

Speaker 2: Thank you, Kelly.

Speaker 1: Ivan Serrini with Farooq Security.