Alan Shimel welcome to the Noise 2 Signal podcast. It's such an honor to have you on. Welcome, Alan.
In the honors mind, my friend Mehul who would have thought you'd be turning the tables on me here, putting me on the other side of the camera? But I'm happy to be here.
Yeah.
Alan, you are in some you are in some respects a kingmaker in cybersecurity media. But if you look back at your history, you started as a law student, you have a JD from New York Lice Law School, and then you ended up becoming the chief strategy officer for Still Secure, competing with the Tenables and the Qualys's of the World. And then, fast forward, you are now like a tech cybersecurity, a media tech magnet, right? So I want I want to start the interview there.
How did a law student become like a key person in cybersecurity media?
Well, you know, the easy answer is if you live long enough. the actual answer is a little bit more involved. You know, I grew up in a time, Mahou, where quite frankly, taking computer classes in undergrad, where it was still key punch cards. Yeah, punch cards. I remember in law school, LexisNexis had just come out.
And the idea of going to a computer to do legal research was revolutionary. mean, because before that, you had to go through it. We've all seen the law library books, those thick books by Shepard and Blackstone and everything. And you'd have to do research by manually looking stuff up. It was impossible, not impossible, but it was very time consuming. LexisNexis changed.
the way lawyers did research, the way he wrote briefs and so forth. And to me, that was the first inkling of just how powerful computers could be to even old kind of conservative professions like law, or say medicine, finance, et cetera. But I went on with my education, I passed the bar.
23 years old and start practicing law. Funny thing about lawyers, we never used Microsoft Word for the most part. Lawyers used WordPerfect. For whatever reason, WordPerfect became the word processor of lawyers, for lawyers. So we had WordPerfect in my office, know, my firm And we had one guy who was like a part-time IT guy, right? We running this
This is before Windows 3.1. People are on DOS and we're running Novell NetWare. And you know, it was a different world. But I became the lawyer expert on WordPerfect. So all the other lawyers, when they had a question about how to do something, I became the go-to WordPerfect.
Were were you the tech savvy lawyer of that era?
Yeah, I want but you know what they say, the one-eyed jack is king in the land of the blind. So I was far from tech savvy, but I knew word perfect. That led to me actually creating like I started bringing computers home. bought laptops, know, desktops. didn't have laptops. These were 386 machines. I'll never forget the first ones I bought my house with Zenith.
That is great.
386 machines. First I got one, then I got another. Now you have two computers, you've got to make them talk to each other. So I started learning a little bit about Novell Netware and networking and everything. And I was very fortunate that that's when TCP-IP, the TCP stack, started catching on. And I recognized that that was a hell of a lot easier than the Novell stuff, to tell you the truth.
especially for a home network like I was building. And also I guess around that same time, Windows 3.0, 3.1 came out maybe, you know, in the late 80s and it didn't have TCP built into it, you know, it didn't have the stack built in. had to add.
You had to install I believe you had to install a network card and install the software.
Yeah, well you had to put a network card, you had to install the software stack too. It didn't have it built, you know, ethernet or whatever.
Yeah to install the soft.
Th you know, the thing that I find the most funny in this entire episode is to be called a tech savvy, you just needed to know word perfect. Like if you knew word perfect, you that's all like no.
Yeah, you are a guru. But like I said, the one-eyed jack is king in the land of the blind.
I mean, if you look at if you look at the tax tech stack now, you have to know the cloud, you have to know the models, you have to know the harness, you know, to be like to be like, you know, to get to be calling called a tech savvy in a it but you know, how did that transform into becoming like the chief strategy officer for a cybersecurity firm?
Where to start? Yeah.
So,
So there were a couple stops along the way, you may not know this. So what happened was I created a lab at my house. And one of my best friends who I knew since we were two and three years old would come over and we would play computers. Back then, the internet, we had wide area search, Gopher for, no web yet. But we also had things like GE's Genie. Now I don't know if you're old enough to remember this.
Interesting.
No.
Genie, CompuServe, there were networks, there was the well, were the message boards and stuff. So I became a bit of a computer nerd, right? More than a bit. It was my passion, my hobby. I went to work for a living as a lawyer, but I'd come home and play computers for as long as I can with everything.
Did you end up ever practicing law or were you were you practicing
you know, I practiced law for five, six years.
And then and you always had the incl inclination towards tech, I think. I think is that
The computers were my hobby, my passion. Law was my profession. And what happened was the web came out. The web came out in the mid 90s, 1994, 1995. Netscape came out, I think, 1995, something like that, in beta. And that ignited my curiosity and passion and imagination.
Mm.
started making websites for fun for my friends and myself. The next thing you know, me and my friends would play in my lab, making websites for everyone. We started charging people. Then, and here's the switch. Here was where the switch went off. One day I came to my friend and said, you know, this making websites fun, but I'm tired of people saying they don't like this shade of red or that shade of blue. Let's just.
Hm.
store their websites. We'll be like digital landlords. If we need more space, we'll buy more hard drives.
Mm.
And you know, we used to go down to this one magazines are still red magazines. didn't read them on the web. We'd still go down to the store and wait for PC Mac to come out every month. A computer shopper, all these big magazines. I took an eighth of a page ad in the back of PC Mac. I'll store your website for $49.95 a month.
In 18 months, we were storing, we didn't even call it hosting yet. We were storing 5,000 websites. And 18 months later, I 50 bucks, I had $50 a month. So do the math. I went to my wife, we didn't have children yet. I said, Bonnie, I can make as much money as I'm making with more. And I love this. I feel like I'm not working. It's my passion.
you know, I want to do this. And so I made the move. The company was called TriStar Web Creations. At the time, New York City was just opening up with the recording Silicon Alley, like Silicon Valley in New York. Silicon Alley was downtown. We took an office there. We opened a real business. And as I said, we had 5,000 in 18 months. We kept growing.
Yeah.
wow.
In 1997, I actually sold the company to a company that was doing a roll-up of hosting at the time called Sage Networks run by guy named Brad Feld. Brad is a co-founder of Techstars at the time he was managing partner, SoftBank Venture Capital. And we sold that company late 97, early 98. I went to work with Brad and the company. That company became Interline.
We were one of the largest hosting and then ASP, application service providers. We went public during the dot com. I, by the way, was VP of biz dev and then got into corp dev. We did 30 acquisitions in 36 months. So I got a tremendous amount of experience, right, in the early dot com. Plus there were 250 companies in the soft bank, Coretzo, that I dealt with because they all, you we were a large.
Wow.
player in a public company, the IPO. And so they all wanted to do deals with us. A ton of biz debt stuff, a ton of corp debt. It was a tremendous opportunity to get a business education on top of my law background. And by the time, then the dot com bubble burst and I left Interliant, I guess in 2001, to form GoSecure.
became still secure. But by then I was an internet business person, mean, things I learned from Brad and the things I learned from that experience have powered my career ever since. So that's where all that comes from. Now I made a very good, I'm sorry, ahead.
Why why k why still secure though? Like why go from like a web hosting web hosting in
I'll tell you what. So the day I closed selling TriStar to what became Interlion, my friend Brad Feld said, let your friends run TriStar as part of Sage Network. So I need you to come work with me. I want you to go to Boston next week. And there's a kid there that he's on his mentor.
He's a really smart kid. name is Raj. Raj Rajat Bhagava. I want you to go up, meet Raj, and you and Raj are going to work together. So that Monday, I flew to Boston, and I met my friend Raj Bhagava. Raj had just graduated MIT a year ago. And he had started a company called NetGenesis, also went public in the dot com. He wrote the business plan for what became the internet.
Raj and I became really good friends. He was much younger than me, but Raj was probably the smartest person I ever met. he, Raj has now done eight or nine startups. He's currently the CEO founder of JumpCloud, a pretty well-known company. But anyway, Raj came up with the idea when the internet was really, like data centers were really proliferating and you know,
Mm.
It wasn't cloud yet, but it was getting there. He said, we're going to have distributed applications that need to be managed. And I helped him write a business plan for a company called Lattice Networks, which was going to manage distributed workloads way, way early.
Mm.
That was in 2000. In 2001, he called me and said, you know, this lattice networks isn't going so good. We're going to have to close it. But there's a guy here, we worked with him in a line, he wrote something for what we call intrusion detection prevention. What do you know about security? Said, I know manage firewalls. I know a little bit about it. He said, well, can you take a look at this? me what you think. A month later, we launched those security.
We actually pivoted Gladys Networks, which was always the official name, into Still Secure. And Raj was CEO. I was Biz Dev and Strategy. And Mitchell Ashley, Mitch Ashley, who still works with me today, he's now a future and as an analyst. But Mitch was here in Tech Strong with me and other companies I've done. Mitch was VP of Engineering, CTO. And we built Still Secure.
Yeah.
This is 2001, security games are very different games.
Yeah.
And you know, interestingly, that is the time when some of these vulnerability management players also came online. The Tenable came online, Qualys came online, Qualys came in with a very different point of view. It was like built on the internet. It was kind of like cloud one point Like in even in that that it was crazy. People laughed at Philippe. If I don't if I don't know if you remember it, people laughed at him.
Yes.
It was crazy. was crazy. No, they did. Well, that's how I used to sell against it. Right? We would go in. So still secure. had an intrusion prevention product. I forget the name of it now. Not border guard. Strataguard. Then we had VAM, Vulnerability Assessor to Management. And then Safe Access was in that network access control.
But back then, vulnerability management, Qualys and Tenable weren't the first. Quite frankly, Foundstone was out there. Foundstone, EI, market at EI. But then you had Philippe, who had come over from big enterprise software, Borland and everything like that, and started Qualys. You had Ron and Renaud getting together, forming Tenable. We used to sell against Qualys by saying,
Yeah. Yeah, yeah, Fountainstone was.
Yeah. Tenable. Yeah.
Are you going to trust your vulnerability data up in somebody else's data center? Up where, you know, we point up and say up there, no one knows what up there is. That's where your vulnerability data is going to live. Are you crazy?
I'm I'm curious. I now that you mentioned I'm curious, were was the term the cloud relevant in those the in that era or what was it?
No, we didn't use the cloud. The cloud didn't come to about 2005. This is 2001, 2002. It was in Qualys's data center. Qualys used to send you, you know, and the other thing about Qualys that they pioneered was they used to call it like the hacker's eye view. So it was an external scan, right? Where everyone else was selling pizza boxes, because back then you didn't just sell security software. You sold it as appliances.
Yeah.
Yeah. External, yes.
Mm-hmm.
So you sold the pizza box that stayed inside to scan your network.
And those were very difficult to deploy, difficult to scale. I mean, I was at Tenable in those days, and you know, it was very difficult. We would have like a sales engineering team for onboarding a new customer. your RPA, this this package, this exe doesn't work. let me send my guy over. And but then Philippe saw something different in the sense that he said, This is all crap. I don't want to deal with this anymore. I'm just going to deploy this in my cloud. And and Philippe, for what it's worth, he was an interesting character.
He was an amazing character. He was. I had the I had the, you know, the I was fortunate enough to meet with him and work with him for like two years. And he he was in his seventies, but he still had the energy of the youth. You could see the passion in him every day. Every day.
Philippe was a wild man. He was a wild man. I mean...
So when I met Felipe, he was much younger. He was probably in his late 50s, early 60s, mid 50s. And he was a dynamo. In many ways, what he pioneered in terms of what we call cloud today and everything else, external scanning, risk management, so much of the vulnerability space Felipe had in his brain.
Yeah.
Yeah. Yeah.
But not only that, I mean, look, we made a living of hiring ex-Qualis people. Because the thing about Qualis with Philippe is it was a bit of revolving door there. People would stay a certain, you probably saw it, you were there two years, that's a long time. No, years, that's a real long time to be a Qualis. Yeah.
Yeah. Yeah. I was there for four years. I was there for four years. So you know it it yeah, it prepares you for life, I would say. I can do any job. You put me in the presidency, I'll take it.
So, you know, it's interesting though. When I competed against him, I was his competitor and we were fierce competitors. Philippe was a fierce, fierce competitor.
You know, the thing I remember is his earnings calls were wild. He if you listen to his earnings call for college, those were wild. He would just he he was
Yes. He said what he said, he said what he thought. But you know the thing about Philippe, so I didn't work for him, so I didn't have that relationship. When the day was done and you could sit down and have a drink, have a conversation about the state of the industry, shop talk, you know.
nicest gentleman quality brilliant person you'll meet at Barnard.
I agree. I mean I I I had the same experience. I mean, he was feared, he was feared as well as loved. Yeah, in the you know, with equal amount.
Yes. And that makes for a great entrepreneur. It makes for a great founder. And it has a lot to do with his success. Once I, yeah, well, he did. And he wasn't shy about his point of views. And that's okay. I'm from New York. I'm not shy about it either. we, once I left Still Secure, and we were no longer that competitor, we became much friendlier.
Yeah. He yeah, he had a point of view. He had a point of view.
And so I'd always make up my business if we were at RSA or another show or if I was in the neck of the woods there to try to grab a bite with Philippe or at least stop in to say hello and chat a bit. Because I respected him. I respected him a lot.
Alan, so let's switch topics and let's bring the focus back to you. Right. So you did you did the you did this still secure thing, you competed with all the legends of the industry. But then next thing we know is you are like a media tycoon. Like you are running dev DevOps.com, Techstrong TV, all these different media properties. How did you become like a cybersecurity media magnet? Like what what was that journey like?
Okay.
You know, I'm the accidental entrepreneur, but I don't know. Look, this I do know. When I got into hosting, what became hosting with TriStar, I became pretty well known in the hosting industry. I'm not shy. make my way known. When I went on at Interliant and we were ASP in internet, I made a lot of, I made a great network there. When I'd still secure, I really became
very well known because it was a different, the security industry was you could fit it in your hand, right? It was much smaller. So I became pretty well known there. It was also an interesting time and my friend Brad Feld, again, was instrumental. Brad came to me and said, hey, you got to stop blogging. Everybody's blogging. This is 2003, 2004. Everybody's blogging. I okay, I'll stop blogging. From blogging podcast.
Then Brad came to me one day and said, Alan, I want you to meet my friend Dick Costello. Dick has a, you know, Dick from Twitter probably, was the CEO. Yeah, was, this was before Twitter. He was the CEO of Feedburner, a company that was trying to normalize RSS feeds. He said, Dick has this great company, Feedburner, they're normalizing everybody's feed, but he can't make any money. He wants to make a network.
Wow. From Twitter, from the from Twitter, yes. Yeah.
He wants to create networks of blogs that have a common theme. Brad said, I'm going to do a VC bloggers network. Would you do a security bloggers network for Dick? I said, sure. Happy to help, So I put together what became the security bloggers network. We had our first meeting at RSA in 2005, maybe 20 of us, 25, 30 of us. had drinks. We talked about it. But I put together this network.
Interesting.
Six months later, Dick Costello sold Feedburner to Google for $90 million with zero revenue. Google bought it just because they wanted the RSS fees. Dick went on to be the CEO of Twitter. Six months after that, Google calls me and says, hey, we got your name from Dick Costello. We got this Security Vloggers Network thing. We don't know what to do with it. Would you please take it back? I said, sure. So I took that back.
Yes.
For the next 10 years, we had huge events every year at RSA, as well normally at Black Hat, where the bloggers network, the bloggers, and the thing was these people, came up with new security, but they were quickly becoming CSOs and CSOs and CEOs of who's who of security. And our bloggers networks parties were legendary. had blogger awards and
So I became instrumental in or a key person in this whole security media world, right? was on, you we all knew each other and we all wrote about the same things and we went back and forth.
When I left Still Secure in 2008 and 2009, or 2009, 2010, I still kept logging. I still kept writing about security. kept talking. Oh yeah, I write, I write religiously. Yeah, but I started speaking and doing everything else. And then security besides Las Vegas. It's funny. Next week is Black Hat.
You still do. I mean you're still active. You're still active. Yeah, you're still active, very active.
B-Sides Las Vegas is still there too, doing a great job. It's a great event. I went to the very first B-Sides Las Vegas. The next year they asked if I would be the sponsor running it. You know, I'm responsible for getting sponsorships. Sure, I'll help the community. And I did that. It was a great B-Sides. And another guy who was on the board with me that year for B-Sides was a guy named Gene Kim.
probably heard of Gene Kidd. Gene was the founder of Tripwire. Yes, he wrote Phoenix Project. So after, you know, I live in history, Mehul that night after besides Vegas, Gene says, hey, you want to grab dinner? I said, sure. I went to dinner with Gene and we had about two bottles of wine over dinner and he takes out a manuscript of a book he's worked.
Gene K yeah the the book he yeah he wrote the book the Phoenix project, yes.
which was the Phoenix project, and says, let me tell you about DevOps. And Gene tells me about DevOps. And I said, what a great thing for security, right? Because I'm still looking at the world through my security glasses. This is going to be great for security. We got to do DevOps. And that becomes my new passion. And I start writing and speaking about DevOps and security. Crazy how this works, right?
Interesting.
Brad Feld calls me. He says, hey, I'm doing, well, you should. Brad's a legend. I mean, he's written all kinds of stuff. He's started so many, he's a BC, one of the well-known BCs, just a great guy. Brad says, by now Brad has started Foundry Group, which is his latest fund and stuff. He Maldives. He said, hey,
I have to speak to this brad guy now.
Mm.
We're doing a DevOps primer for my Foundry and Techstars portfolios. Would you come out and do and talk DevOps to them? I said, sure. So I got to Boulder. I give a presentation on what DevOps is, security and everything. Afterwards, Brad says, we needed a site about DevOps. There's no central site. He said, you're the perfect person to do it. Why don't you go do a site on DevOps? I said, that's a good idea.
I look into it, I say, I may need some money. I'm come to you to raise money. He said, put it together and we'll see. Put a business plan together. I put a business plan together for something I call DevOps Central. I had that done later.
Then I met the guy who owned the domaindevops.com, guy named Martin Modin, great guy. At the time he was working for Amazon, Martin went on to do amazing things. I talked Martin into being my partner and giving me the domaindevops.com. And I never raised money from Brad because IBM signed on immediately, CloudBee signed on, AppDynamics, few others.
And so was born DevOps.com, the official corporate name was MediaOps, doing big pieces of DevOps. Once DevOps.com took off from the hole, my heart was still in security. I went back to my security bloggers network and started Security Boulevard. And everything else from there is history.
Mehul (26:48.224) You know You know the thing that I that fascinates me about your b behind your story is I see the same patterns that you saw in nineteen ninety five repeating all over in twenty twenty five. Let me explain. So you tell me that I know I I had a passion for technology. I started building websites, I would, you know, hire I I would go and build websites. Now people are doing the same thing with Cloud. And people are essentially building websites with Cloud, deploying them.
yeah.
Because you know, not many people are tech tech savvy. So they can now they can so you know, that is one pattern that I'm seeing. The second pattern that I'm seeing is everyone wants to there is always this need to find a way to get the information out in the right way in the right channels, right? So your your the security readers blog, you know, that was an outlet to talk about the latest integrated but there was no central there was no central place for it to go. So it essentially landed up in
No.
all these feeds and then these feeds became like aggregators of the twit they became the Twitters and the Reddits and the you know all none of them right there are so and now going back to the DevOps is it was the combination of dev developers and operations like combining developer and operations but now if you look at the latest generation of it it is combining product engineering and UX. Like now there is because now you know because the product
There was none of that then, right? Yeah.
The product guy can build the product, the engineer can design the product, the designer can b you know write the code. So now there is a merging of all these different facets of the industry merging, similar to DevOps, but in a different in a in a different spectrum. In a in a broader scale because security is part of it.
in a broader sense though too, because security is part of that as well. A lot of this, they've given the name platform engineering to a lot.
Platforming, I mean, it's very difficult to know like, you know, for when DevO and Ops came together, like the the, you know, they were combining these two fields into one. And like you should be, you should be able to do both Dev and Ops. You should just write the code in one time, you know, infrastructure as code and whatnot. The question I have for you is you've seen this patterns over and over again, and you and you see different founders coming up the journey in every era of industry.
Have you seen any patterns in the storytelling? Because one thing that I have realized is you have to tell the story really well for somebody to understand what it is that you're selling, right? Because not everyone is adept at telling the story or at putting a grand vision behind it. You could just say, Well, I'm just selling RSS feeds, and that is a very crappy story. But then you you could you could tell I'm building Twitter, that's a big, much bigger, you you know.
It is becoming it becomes a much bigger story than whatever it is. I'm curious, given all the things that you've seen, and especially with your seat now, you're essentially the editor-in-chief over DevOps.com, TextRong TV. You see so many founders come to you, they talk to you, you interview them. And this is kind of the thing I wanted to tell you in the start at the start of the interview. This is like the glitch in the matrix where you interview thousands of people, you know, over a year.
I haven't seen an in-depth interview of you because you have all these insights. Like you have lived through this, you have built so many companies. And so I'm fascinated to just interview you because you know all of this rich history that you're bringing it to the table, because you've seen these patterns, like whatever you see in the cloud today and AI today, you've seen that in like in 1995, right? So my my question to you is like, how has storytelling changed? Or is it essentially the same?
With AI.
I I get what you're asking. So you know, I think storytelling is the second oldest profession in humanity. I think humans were telling stories before we had written language.
What's the first? What's the first?
Well, I don't want to get into it, but look it up, it'll tell you. But it's one of the oldest professions. Let's leave it at that. But humans have been telling stories for as long as we've been human. The technology we use to tell the stories have changed. We went from strictly oral to writing on cave walls.
Yeah.
Yeah.
on papyrus, stones to papyrus, to paper, to printing, to electronic like we have today.
As someone who's been writing, know, I mean law school teachers who write a lot. Yeah, and concise and, you know, make your arguments and so forth. Nothing holds a candle to what's possible today with AI. I know a lot of people say, oh, you can't, it looks like it's AI, it sounds like it's AI. Yeah.
Yeah, write well. And write well.
Yeah.
But every new release gets better and better and better. And I think this is going to be humanity's greatest storytelling epic era ever. Because I think our ability to tell a story now is 100 times more than it was, let's say, when I first started writing.
But I are you seeing that change in the C levels and the executives and the startups of how they are telling the stories or is it essentially the same or they all subset up
Bye.
Yes and no. You know what, we're in a learning phase right now, right? So, you know, this goes to the heart of the whole AI question. Is AI going to replace the storyteller or is AI going to make the storyteller better? Is AI going to replace the developer or make the developer better? Is AI going to replace the lawyer?
I I choose a second half.
or make the lawyer better. This is where we are right now in this AI story. I believe it's going to make us better, not replace us. And I think that's the key.
Totally agree. It replaces the tasks, not the jobs. So it replaces like you know, like you're if you're an accountant, you probably used paper and pencil in 1980s to do the accounts, but now you use Excel to do those accounts, right? So you know the job still the job is still there. Exactly. But you just do your job better, you do it do your job better, much more accurate, less errors, and and there are more accountants and more effective.
Exactly.
Yeah. Yes, exactly. But Excel made more accounts, right? Not less accounts.
and more effective. And that's the key. That's the key.
Alan, I want to for the next topic, I want to get into you are somebody who has a lot of points of views. so I want to get I want to get you know some of your hot takes, especially since we are talking about AI. I want to get some of your hot takes. you know, one of the things AI enables you to do is to build faster. Right? You can build faster, you can prototype faster. And the comment that I read, one of your comment that I read, is you know,
Okay.
You don't want to be building the wrong thing faster with AR.
No, you know, so, but again, nothing is new under the world. We recycle stuff. I was always taught, get to know as fast as you can. If you're gonna get to know, get to know as fast as you can. Whether we're talking about selling a product, buying a company, or whether a technology solution works, you wanna get to know as fast as you can.
Mm.
Like don't delete that. If it's no, it's no and move on to the next thing. And so it's the same thing here with AI, right? You got to get to know as fast as you
So I mean so are you basically saying that you should fail faster, try many things, figure out where you know
And that's a silicon, you know, look, Silicon Valley, that became a cliche. Fail fast.
Yeah, feel faster. Yeah.
fail fast is a lot, know, flip side of that is you've got to break a few eggs to make an omelet, right? And so you've got to accept the fact that in making omelets, you're breaking eggs and some eggs get broken and can't use them. And that happens.
And on the on the topic of Silicon Valley, Silicon Valley has in has been at the at the forefront of AI. They've been building this technology, they have been driving this technology forward over the last two or three years. But I'm also sensing that they are now getting to the point where they're trying to limit the competition around AI. I'm there with the just last week there was a big bruha on open source versus closed source. There was an effort to ban
or at least limit the open source models in favor of some of the closed mo closed source models. what's your take on what's your take on open source versus closed source AI?
So I've written a bunch of articles since last week on this subject. So first of all, let's be clear though. It's not open source, it's open weight. And there is a difference, right? These models are open weight models. There's a couple of things that just jumped out at me. First of all, in what bizarro universe would we think that the Chinese Communist Party is the champion of open, correct? In a country that,
Open ways, yes. Yes.
that has the great firewall of Chyter that restricts what their people can do, they're out here championing OpenAI, not the company OpenAI, but open source or open-weight AI. That should tell you all you need to know because, and we here in the US, we've become the people who are saying, no, no, we're gonna firewall it, we're gonna lock it down, we're not gonna be free.
Yeah.
And there are also these false narratives that the you know the data is getting siphoned to China because these open rate models are hosted here. No, no, no. Not only that.
distillation, call it. They may very well be distilling it. But again, the irony here, let's look at this. Anthropic and OpenAI are mad because they're accusing China of stealing information that these Chinese companies are using to build their models from the very same people who stole everything on the internet to build their model.
Correct. Yeah, they distilled these models from the open internet. They didn't ask your security readers blogs on permission.
Yes, no, they didn't. probably use my content on DevOps.com and the Security Boulevard and everywhere else. no, you know, and I'm not the New York Times to go sue them and get paid for it. so there's a certain irony there. Let it not be lost. However, here's my, I'm a big, you know what, Ron Gula, who another friend of yours.
And never asked you never asked you ever.
Yes.
Ron Gula taught me in 2006 or so, open source, you free as in freedom and free as in beer. A lot of people go to open source because it's free beer. know, theoretically, you're not paying for the software. But the real key to open source is free as in freedom. And freedom is not something you can clamp down. The tighter you squeeze, the more it squeezes out.
And so that's why I'm a big supporter of this OpenAI initiative. know, Jen Simuang and the whole bunch of companies have signed on trying to get us here at Futurum to sign on. Because I do believe open is the future. don't know if I told you this, I'm finishing a book right now, Final Edits, covers done, hopefully in a month or month and a half, called The Indispensability Trust.
It's about AI. It's about, but it looks at historically things like the railroads, the electricity grid, the phone grid, the fiber optic grid of the dot com era and how the amount of money that got poured into those grids, but where the lasting value was made. The same thing holds true. That pattern is going to repeat itself in AI. All the money we're pouring into
data centers, $8 trillion. And the models themselves, they're going to be commoditized whether we like it or not. We could either commoditize it ourselves and have open AI here, or we're going to let the Chinese do it. But it's going to be open, it's going to be commoditized, because that's not where the real value lies. The value lies up stack. What are you building with that AI?
Mm-hmm.
So you so you do you think the value is not in the models but is in the applications like the cloud codes of the world and like chat GPT?
Well, but look what Claude's building Claude Design Claude Health Claude Law Claude and that's just the beginning. I don't think we've seen the killer app of AI yet. Right. I think we're still learning what we're going to build with this AI. Yep. So that to me, so this fight over trying to keep opening, you know, open weight AIs out is a losing fight.
Interesting.
You're experimenting, yeah.
It's the right way to go because the models will be commoditized. They have to be. Because otherwise, if it's not commodity, if it's not cheap enough, it'll never be used enough to do what we want it or we think it can do.
I'm curious, like you know, with these models becoming open weights, they also have a tendency to encode their their own ideologies in the in the models, the ideologies in the models. I'm curious what do you think? And this, you know, encoding ideologies is always a risk for the governments. They have always w certain ideologies to be promoted, certain ideologies to be demoted, and so on. Do you think that would give rise to sovereign AI or s or sovereign AI?
I think we already have sovereign AI. Well, it's not tendencies. The Chinese government, the Chinese Communist Party finances a lot of these open AI companies. the open weight, open weight. Right, what are the open weight?
open source companies, open open source or open weight. Yeah. Open weight models. Yes, yes. And there I mean there are there is some and there is some ulterior motives in the sense that they it helps them like take
They have told them you cannot put these things, they put guardrails around, go type in Tiananmen Square and see what it says, right? These kinds of things. So, you know, that's a political thing that the Chinese are doing and they're the Chinese, that's what they do. But to me, you know, right now we have this China-US thing for supremacy. What I think...
You know, that's exactly what I'm saying. Yeah.
Mm.
Yeah.
The lesson there is though that it's not, you don't need the most powerful AI to do every job, right? Most jobs can be done with less than the latest, greatest model. And I think you've got to look to Europe. You've got to look to India. You've got to look to the other places in this world that have technology. Asia, Singapore, you know, these places, that.
Mm, yes.
they're gonna be the ones who choose what AIs they want and whether they want, you know, they all wanna be sovereign. Because what they're all afraid of is that whether it's the US or China, they're gonna build their ideology and they're gonna keep control, right? They're afraid that, you know, they can't wake up one day and the administration of Washington decides to shut something down, right?
Into it. Correct.
Not only that, I mean there is also the there is also the risk you know, your younger generation asks questions about some topic and they get a completely different answer from a Chinese model versus an, you know, American model or some other European model. And then it's kind of like the Wikipedia, like what is the source of truth at that point? Because then you can program the young minds and go from there too.
Yeah.
So this could be a whole different podcast, right? I mean, why is that any different than social media today? Isn't that the problem we have in a wider scale, not just with young minds, but with everyone's minds? No one knows what to believe. No one knows what's fake news. No one knows what's real. And I hate to say it, but I think parts of our government sow this. When we say this is fake news and that's fake news,
Correct.
To a lot of people, it's obviously not fake, they can tell. But just by sowing that doubt in people's mind, you put everything in question, right? And this is...
Yeah, I mean it's I mean I'm very skeptical of what I see on the internet now, especially. I mean some of these AI photos, the video, the voice, the video, the voice, the image, everything looks like you know
We all are. We all are.
Yeah, no, it's something we as a species have to come to grips with. How are we going to get ourselves out of that mess? It's a mess.
Alan, you've been in the media business for almost ten, twenty, twenty, thirty years, right? I'm curious when these AI models are so good at writing copies, summarizing text, you know, in the in some ways writing really good, well-structured blogs. at some point I think they'll also get to the point where they can create videos like Hollywood quality video.
The area DR.
Th and the voices, some of the times, you know, I use it in my podcast to do the voice for the for the voice of God and so on. What do you think is the future of IT or tech journalism with all these tools coming to c coming to for for somebody like you?
Yep. Yeah. So a couple of things, a couple of thoughts. And again, I've written a lot on this particular subject. Number one, AI doesn't replace good writers. AI makes good writers better. Right? So you're always going to have good writers, good journalists, and they'll use AI for the tool that it is to make their writing better and to do more of it.
Hmm.
Right, I use AI a lot, right? I argue with my AI to kind of reason out my points and thoughts and everything else. I ask it to edit my articles, you know, help me with it. I think the bigger issue facing tech journalism though, is the demise of the open web, right? The demise of the open web. All media, not just tech media, in the age of the internet.
Interesting. What do you mean?
was built around Google. Google was a central piece of it because what happened was there was a covenant made. Media let Google index all of our sites. Google crawlers, you opened up your electric crawlers index room. Google then amassed that and you and me and everyone else searched in Google. And when you searched in Google,
Yeah.
Google came up with the links to the right sources for your question. And you would click whichever link made sense to you. journalists or media companies generally got anywhere from 60 to 85 % of their traffic from what we call organic search, Google. Google was 99 % of
And that was the, that powers the open web. That powers the views you get, the advertising, everything. Google made a fortune. Google became one of the most powerful, richest companies ever based on that, because they sold advertising around it and placement, the whole SEO thing and page search and all of that. AI broke the covenant. Google now
follows what they call a zero-click policy. So when you do a search on Google, the majority of them are returning an AI answer, an AI summary. That summary is made up of my information, your information, all of our information. But Google now is taking that information.
Mm.
Yeah, summary. Yeah, it gives any ice.
Yeah.
and keeping the visitor at Google, not sending them, they're not giving out the click. They're keeping them there. And so most media outlets have suffered anywhere from 35 to 70 % less clicks from Google because Google is keeping the traffic.
What's the what's the solution for this? Is this solution to stay stop crawling?
Well, I wrote an article this week that several major mainstream media, People Magazine, I think I was the Washington Post, major newspapers are talking about shutting Google off from indexing their sites so that it no longer has the information to serve up in AI.
Would you be would you be open to it if Google paid for it? Like to index DevOps.com? Would that be a solution?
If Google wanted to pay for it, well, let's be clear. Open AI, 30, 35 % of searches now will be done in Open AI and CORD or other AI, know, via AI versus Google. And they're paying, you know, well, Open AI and CORD are paying to use that information to train their model.
Yeah.
Yeah, but that is only for like the premier website. It's not paying everybody. It's probably paying New York Times and Washington. Yes. Not like DevOps dot com and
No, no, they're paying like the New York Times and stuff like that. No, and you know, we have 50,000 DevOps.com articles that I'm sure sucked in there. I think the answer, I don't know what the answer is, to be quite honest with you. I don't think Google, I think Google has to serve up the answers in AI, because they got to compete with OpenAI and cloud.
Hmm.
But that being said, it's a real issue. It's a real issue. And the covenants broke. And the thing is, no one or even no three media companies are going to be able to break this. There has to be a media-wide effort to get Google to change its ways or to do something different.
So is the outcome of this fewer media companies in the future if Google is allowed to do what it does? And
Perhaps, maybe, but no, I think the better answer is media companies have to change their model. They can't base their model on getting their traffic from Google. They've got to work harder to get a dedicated readership, a passionate readership who signs up for their newsletters and their learning events.
Yeah. Yeah.
I'm not saying they got to pay, that's maybe something down the road, a subscription model, like the Times or the Post or Bloomberg does. But at the very least, expect to see more media behind registration walls, pay walls. what happened, because what Google did redefines the relationship between Google and the media publisher.
Mm.
The relationship between the media publisher and the end user has to change. The relationship between the media publisher and the sponsor has to change. You know this. You now have a company. You have a PR agency. They promise you earned media, free media interviews. We can't afford to do them very much because we've got to do things that keep the lights on here.
asked it.
Instead of earned media, we offer paid media and other ways of getting the word out for you that allow us to have a business. Right? Because we can't just count on traffic from Google.
Alan, last question, last question for today. What you've been in this industry for over 30 years. You have seen so much change. And there is there is a lot changing in IT as it exists. IT and security. what do you think is the future of IT? What is the future of jobs in this world that is coming our way? Because the way systems are deployed, systems are secured.
It's completely changing. What's the future from your point of view, your vantage point?
Yeah. So first of all, let me read off this.
A frustrated lawyer, computer nerd has been able to spend 30, 35 years doing this. And I always tell my kids, if you do what you love, you never work a day in your life. So I've spent 30, 35 years just doing this, and I love it. It's not going to go away. It's going to change like it's always changed. Change is constant. You know there's a saying in tech.
If you're not moving, if you ain't moving forward, you're dying. Right? And it's always going to be moving forward. But there'll always be place for people who are smart, who know how to apply the technology, use the tools, have the imagination. That's the biggest thing. AI doesn't have imagination. I don't know if it ever will. Humans have imagination.
Yeah.
Yeah.
Yeah.
And if you have the imagination to dream about what can be, and you want to use AI or whatever comes next to help you make that dream come true, that'll always be true. That'll always be there.
Alan, that's a great note to end the interview. Thank you for coming on Noise to Signal. It's an it's an honor to have you here. I think you have you really enriched the entire Noch to Pugnal Noch to Signal podcast series. So I'm super grateful. Thank you.
My pleasure, man. I'm glad we were able to do it. You we had you on TechStrux TV. I wish you lots of luck. Anytime, anytime, my friend. We'll see you maybe at the Black Hat. Alrighty.
I'll stop I'll I'll stop it here.