Realists At The Table: How To See Through The Hype
With Chris Romeo, Izar Tarandach, and Matt Coles
Has cybersecurity traded curiosity for the promise of a paycheck, or are experienced practitioners simply seeing another generation's version of the same hype? Chris Romeo, Izar Tarandach, and Matt Coles examine the industry's changing stereotypes, from hoodie-clad specialists to ambitious founders.
Episode chapters · 12 chapters
- IntroAudio
- Cybersecurity stereotypes and trustAudio
- Curiosity versus the paycheckAudio
- Founders and the new stereotypeAudio
- Mainframes, cloud, and recurring hype cyclesAudio
- AI, quantum, blockchain, and NFTsAudio
- The Cuckoo's Egg and security storiesAudio
- How hacker movies influenced practitionersAudio
- Threat modeling moves beyond a nicheAudio
- The search for shortcutsAudio
- Garbage in, garbage outAudio
- AI personality and closing thoughtsAudio
About this episode
Has cybersecurity traded curiosity for the promise of a paycheck, or are experienced practitioners simply seeing another generation's version of the same hype? Chris Romeo, Izar Tarandach, and Matt Coles examine the industry's changing stereotypes, from hoodie-clad specialists to ambitious founders. Mainframes, cloud computing, AI, quantum, and NFTs provide examples of ideas that return wearing new labels. The Cuckoo's Egg and hacker movies lead into a discussion of what made security feel like a community and whether broader adoption changes that appeal. They return to threat modeling, the search for shortcuts, and the garbage-in, garbage-out problem before asking whether AI has a personality worth taking seriously.
Mentioned in this Episode:
➜ The Cuckoo's Egg — Cliff Stoll
Follow AI Security Table:
➜ Home: https://securitytable.ai/
➜ X: https://x.com/AISecurityTable
➜ LinkedIn: https://www.linkedin.com/company/ai-security-table/
➜ YouTube: https://www.youtube.com/@AISecurityTable
Transcript · 38 min conversation
0:09Chris Romeo Hey folks, welcome to another episode of The Security Table. My name's Chris Romeo, joined as always by Matt Coles, Izar Tarandash, and it's another day. The sun is rising and, uh, sun is rising on the cybersecurity community.
0:26Matt Coles Well, on most people, unless you're on the other side of the world, in which case it might, you know, be going down.
0:35Chris Romeo The sun might be descending if you're in Australia right now, where we have a number of listeners.
0:41Matt Coles Woo-hoo!
0:42Chris Romeo Have a good evening, those in Australia right now. Okay, where are you, Matt? You had some ideas about where you wanted to go. I can't even imagine.
0:50Matt Coles I don't know if they're good ideas, but we could certainly try.
0:53Izar Tarandach I'll put my money on 'em.
0:54Chris Romeo Set it up for us. Set the stage. So an interesting question.
1:00Matt Coles And, uh, I kind of wanted to talk about trust today, but, but let's, before we talk about trust, let's talk about how, who we're dealing with here. So in the past, there's sort of been, uh, an understanding. I think there's been a, uh, oh my God, what's the word I'm looking for? Stereotype for the, for the type of people who are in cybersecurity.
1:23Chris Romeo Right.
1:24Matt Coles They are pessimists. They are to some extent—
1:31Izar Tarandach I prefer realists.
1:35Chris Romeo Me too. I prefer to be called a realist sometimes.
1:38Matt Coles Yeah, well, pragmatist is not necessarily the word, but maybe that came eventually because sometimes they also came with a sense of inflated ego. Like I know better, and therefore, um, my— and that's where pragmatism had to come in because that getting shot down so much. Uh, and I was curious what your thoughts were on— what's that? If any of that changed for those, for those who are listening to this and not watching this, Go over to YouTube and watch this.
2:17Izar Tarandach It's the Loch Ness Monster.
2:19Chris Romeo It does appear to be some form of monster attacking that.
2:23Matt Coles Yeah. So, uh, has the profile of individuals changed in terms of who gets into cybersecurity? And to some extent, does, does the introduction of AI change the profile of the individual? And/or does AI have the pro— the personality type to to be in this industry?
2:45Chris Romeo I'm gonna leave the AI piece off to the side for a minute in my initial answer to this, and I'll share something that I've heard. I'm curious if you guys have heard the same thing. So I've been in cybersecurity now for 30-plus years, and so I get the opportunity to sit down with various young people who will say, I'm interested in cybersecurity. And somehow they'll get connected to me. And I'm always happy to take those meetings because it's the next— we're talking about the next generation of what we do here. And we're not going to, you know, people like us aren't going to be around forever. We're going to eventually retire and move to beaches and, you know, wait, we're not old curmudgeon and wait, what do you mean?
3:28Matt Coles We're full of conservatives that we're going to live forever.
3:31Chris Romeo But these are—
3:33Izar Tarandach what do you mean, get tired?
3:35Chris Romeo Oh, yeah, yeah, true. Yeah, yeah, we have been tired.
3:38Matt Coles You have to be tired in this industry. And it's an industry you're always tired.
3:42Chris Romeo You're not tired, you're not doing it right. The point here, here is what I've heard from a number of different young people that are thinking about, they think cybersecurity is something that they're gonna do, or they want to go out and get out and go after, is they think it's a lucrative job opportunity, meaning they think they're gonna make a lot of money. And I've heard it a number of different times. And it just, it's, I mean, did I always wanna make a lot of money throughout my career? Of course. But did I, is that why I showed up to work on Monday morning? No, because I think, and I think we're all in the same boat, like we love doing this. We would do it in some capacity for free, even if we didn't have jobs in it. The money is nice and, and it's good to be able to eat and feed your family and everything, but that wasn't, that wasn't ever my motivation. I was never like, and, and granted, I got started in cybersecurity when it was in its infancy and, but I was never like, ooh, I'm gonna make a lot of money. That was never my motivation. My motivation was we have problems to solve. There are interesting scenarios. It's not the same thing every day. And so this is just something I've heard from meeting with a number of different young people over the last couple of, yeah, maybe year, couple years. And it just seems like it's a different answer than I ever heard, that I heard 10 years ago, 20 years ago.
5:13Izar Tarandach So here's the thing. I think that Matt is right. There was a stereotype in there. And I think that Chris has like put his finger on the stereotype. Perfectly. The three of us, we got into this game by probably making the mistake of turning our hobbies into our profession.
5:37Matt Coles Right? Yes.
5:38Izar Tarandach Yes. And that sort of links with the episode that we had about enmeshment and personality and work and, and all that kind of stuff. Uh, in, in, in the term that, yeah, Chris is right. I, I think that we would be doing this some form of this stuff for free if it didn't pay.
5:57Matt Coles Well, in fact, we do often, right? We volunteer our time, right? With organizations and whatnot that are not necessarily recognized, you know, or monetized, right?
6:09Izar Tarandach And the point is that, and that's great that Matt went there, because we built a form of community around this, I'm not going to say old guard of people who sort of grew up doing this stuff this way, but some sort of, uh, um, common denominator of people, which in turn I think created much of the stereotype, right? But at the same time, At least to me, it's funny. And especially, I understand Chris's experience with young people. I am fortunate to still be allowed to work with many of those, but especially when we go—
7:01Matt Coles sorry, go ahead.
7:03Izar Tarandach Yeah, yeah, I know exactly what went through your head. I'm allowed. I told it as soon as it came out of my mouth. Yeah.
7:11Matt Coles But, uh, he means personality-wise, not legally-wise.
7:14Izar Tarandach Yes, yes, exactly. Hey, Jarve, stay where you are. Stay where you are. Go ahead.
7:17Chris Romeo Go ahead.
7:19Izar Tarandach But the point is that then we go to places like OWASP conferences and stuff like that. And to me, at least, it's almost like, it's almost like it blinks on the radar. Those who come in because they have that same passion that now has been fired by years and years and years of tradition, and those that come in because, hey, it's a great job and, uh, you can do some good money in it and you get to tell people that what they're doing is wrong and they should do it differently.
8:01Matt Coles In other words, reinforcing the stereotype that there, uh, there's an ego and a I know better than you.
8:08Izar Tarandach True, true, true. But on the ego thing, this is going to hurt so much to say, but—
8:16Matt Coles Let's be cathartic. Let it all out, Isar.
8:19Chris Romeo Yeah.
8:20Izar Tarandach I think that we grew up back in the day. Yeah.
8:24Chris Romeo Yeah.
8:25Izar Tarandach That there was some ego going on there and that there was that amazing stuff of, I know arcane arts that you don't. And, you know, I go around with my hoodie up and oh, it looks so cool. Never mind that nobody thinks about it, but I, I'm going to drop those words that are just mythical, you know. Exactly. Fuzzy. Watch as I type very quickly on something with green letters on a black background and oh my God. If only Hollywood could see me now, right?
8:56Matt Coles I know Unix.
8:58Izar Tarandach Yeah, but the thing is, and I am not a dinosaur, just work of some, but the, the, the thing is that, uh, the new stereotype is the, the young founder bringing in the very quick startup that's going to get like millions on VC and either make it big and get bought by someone or just disappear in a puff of snow or smoke. And, uh, even the experience with those founders that the three of us have had of advising these people, or just as they say, giving us 15 minutes of our time to give an opinion on this idea that they have, uh, it goes to show that it's not, I don't think it's, it's ego anymore. I think that we got to a level of maturity. And again, we have talked about this in the past where even the new thing is boring. We, we get, when was the last time that you found yourself getting excited by watching somebody else's product? And it's not that we think that we know better than them. It's just that we've seen this before. We've been there. We worked with it. We tried to, we tried to create it. We, we sell, sold it. We, we, we managed to, we, we did another version. We used a different technology. But, um, it's not ego, it's maturity, I think.
10:34Chris Romeo When you think about, um, maturity, somebody asked me a couple days ago, like, are we in an AI bubble? And I said, of course we're in an AI bubble. This is like my fourth time around of the next big thing, right?
10:53Matt Coles Like Starbuck. Yeah.
10:54Chris Romeo I mean, I remember when when cloud was the, the, the best thing since sliced bread, and we entered a cloud bubble, and eventually that bubble burst, and then everybody just kind of realized, okay, this is how we're going to do business moving forward. But the hype cycle went up, up, up, up, and then cloud wasn't, wasn't the hot thing anymore. And like, how many other things like that have we— we've been through the blockchain hype cycle.
11:22Matt Coles I mean, it's interesting that you say that. Did this all start with the dot-com .com boom and bust.
11:29Izar Tarandach No, once upon a time, electricity was the hot thing until it was just something that everybody has in their homes.
11:36Matt Coles Not for us. I mean, we're not that old.
11:39Izar Tarandach No, no, no, no, no. But what I mean, what I mean is I think that the progression from hot thing to something that's commonplace, like the cloud, is just a profile again of maturity, right? You start with something.
11:53Matt Coles Yeah, I guess what I mean is though, you know, that, that hype cycle really started with the dot-com era, right? For, for us and security in particular.
12:02Izar Tarandach Well, I mean, no, no, no, no, no.
12:03Chris Romeo Remember, remember the first hype cycle was really the mainframe versus personal computer. That's true. Where we were kids.
12:12Matt Coles Yeah.
12:12Chris Romeo Most of us, I think we were all kids when that was happening.
12:15Matt Coles I mean, mainframes, IBM mainframes especially were, were still a thing.
12:18Chris Romeo Right. And yeah, that was the big, that was the big hype of like, okay, we're going, we're leaving the centralized computing. And we're, and we've got this desktop computer that's revolutionizing the world.
12:28Matt Coles Right. At some point we should do an episode on, on this because I mean, as you mentioned, or as, as you didn't mention, but I will now, the hype cycles, things come back around, right? Sure. The notion of centralized computing. Yeah.
12:44Chris Romeo Was a thing. That was a funny one. That was a funny one.
12:46Matt Coles And then it came back around.
12:47Izar Tarandach The network is the computer. Remember the one from Sun?
12:50Chris Romeo That was another. Yeah, that was, that was another hype.
12:53Matt Coles Uh, and now we're coming out of the, now we're coming out of the cloud back to on-prem and local.
12:58Chris Romeo And so it's, yeah, I mean, but I took us in that direction just because part of maturity is we've been around a long time. So like, we're not riding the wave of the AI hype cycle. We, I think we all agree AI is fundamentally changing the world. It's changing how security works. It's changing how development works, but it's also not, we also realize there'll be something else after this. Like this isn't the end of, of hype.
13:26Izar Tarandach Yeah. No, no. We, we already know what's going to, to be after this, right? Because it's been trying to be the next thing for so, so long.
13:34Matt Coles Like quantum.
13:34Izar Tarandach It, it's probably going to be quantum in, in a, in one way or another.
13:38Matt Coles Right. Well, quantum, quantum has an interesting property though that the others don't, right? The other, other hype cycles.
13:45Izar Tarandach It has and it doesn't.
13:46Chris Romeo Yeah.
13:46Matt Coles Well, we're getting a little off topic, but, but being, but it, yes, it does that because of, you know, It is, or it isn't.
13:55Izar Tarandach Or not.
13:55Matt Coles But the thing that quantum will change is those other hype cycles evolve not only the capabilities of the environment, of the systems that we were involved with, and the technologies we had to learn in order to continue to secure them, right? So there's a continuous learning process here. But quantum adds a condition that we didn't have with those other things. And that's that there's an insecurity that was always present that quantum now enables. It's kind of like, it's kind of like before return-oriented programming and living and being able to tap into libraries and use precompiled function, precompiled information to exploit or using exploits for vulnerabilities. You had a pivot problem. You had to get in and break a system, but when return-oriented programming occurred, you could now tap into execution flow and pivot within the system. And that was an evolution of technology that we then had to adapt to. Quantum brings into a situation where all you had to do was passively collect information and now you have an insecurity. and we can't defend against that anymore. We can't defend against it because we didn't defend against it previously. Whereas we can start to defend, we can, when return-oriented programming occurred, I remember being sitting in a DEF CON presentation on return-oriented programming and the one thought came to mind. So if we don't have buffer overflows, does return-oriented programming work? And the answer is, well, yeah, if you, if you close the buffer overflows, you won't have return-oriented programming to worry about. It's like, okay, we can solve that. Well, can't, but we tried. But we can't solve quantum, right? Because if you recorded this whole harvest now, decrypt later problem, it's a different set.
15:59Chris Romeo I mean, information has a value that's, it's on a linear progression from time.
16:07Izar Tarandach It degrades in value and degrades in quality.
16:10Chris Romeo Yeah. So I mean, if you had, if you got information from 30 years ago, it's, it's not worth very much right now if you can decrypt it today.
16:17Matt Coles Uh, it depends on the nature of the, of the content, right? I mean, government secrets as an example, right? Might be a very lucrative now, even if it's 30-year-old tech, uh, fair secrets.
16:28Chris Romeo It'd be like a telegram machine.
16:29Izar Tarandach I'm just looking forward to the day when AI and quantum will get together. So it's going to be the, non-deterministic quality of AI together with the unpredictability of quantum. And we're going to ask a computer, are there any bugs in this code? And it's going to look at us and go, could be, could be. Yeah. I, I could think of some. Yeah, sure.
16:53Matt Coles I might need 10,000 years to calculate that. Yeah.
16:56Chris Romeo I mean, if you're gonna, if hold on, if you're gonna meld together AI and quantum, why don't you just use blockchain? As the database to write to.
17:08Izar Tarandach Oh yeah, no, no, no, that's how you're going to get the output, right? Okay, got it.
17:12Chris Romeo And then we use the cloud. Yeah, that's on cloud.
17:15Izar Tarandach And you get paid in NFTs, you get paid in NFTs. But going back to our original—
17:19Chris Romeo NFTs.
17:21Matt Coles Oh yeah, that was a thing, right?
17:23Chris Romeo Yeah, there's another hype cycle. That was a big hype cycle. People are paying millions and millions of dollars for what were the, some apes, what was the ape?
17:31Matt Coles Oh, the banana things.
17:33Izar Tarandach The smiling ape?
17:33Matt Coles No, the, no, the banana monkeys.
17:35Chris Romeo Monkey must have a name. Had a name.
17:38Izar Tarandach It had a name. It had a name.
17:39Chris Romeo Something apes. I don't know. People paying millions.
17:42Matt Coles And then you had Banksy, right?
17:43Chris Romeo Well, Banksy's pretty famous, is a famous artist. But yeah, he did some NFTs as well. But yeah, all of that's aside. Okay, so we completely went off. Well, we, we followed the path. We fell in a hole. Let's, Matt, pull us back out here, back to your original question.
18:00Matt Coles So back to the personality thing, one thing, the personality and the training profile, I guess to some extent we could talk about the training profile.
18:07Izar Tarandach Oh yeah.
18:07Matt Coles As some, so myself as someone who didn't start in security, I think amongst the three of us, I didn't start my career in cybersecurity.
18:15Chris Romeo I didn't either.
18:16Izar Tarandach I, I thought I was a developer.
18:18Matt Coles I don't remember your origin story. I don't know if I ever listened to your, did you ever do an origin story on your own podcast? Chris on the AppSec Podcast.
18:28Izar Tarandach Oh, now it's the time.
18:31Chris Romeo No, we don't have time.
18:32Izar Tarandach I need to wait. Wait, wait, wait, wait, wait, wait, wait. Hello folks.
18:37Matt Coles Uh, where's Robert when you need him?
18:39Izar Tarandach Chris, tell us about your origin story.
18:41Chris Romeo No, no, no. I mean, the simple version is I started as a sysadmin. Yeah. Okay. Worked for a couple of years as a sysadmin because it's not because I didn't want to be in cybersecurity. It didn't exist. Unless you were in the government. Exactly.
18:55Izar Tarandach It was just work.
18:56Chris Romeo World.
18:57Izar Tarandach It was just work.
18:58Chris Romeo No, I mean, even in, in those days, if you, you had to be somehow attached to the government.
19:02Matt Coles Mm-hmm.
19:02Chris Romeo I'm talking about 1995 or so when I started making my way towards this. And so that was, this admin was kind of the gateway.
19:11Matt Coles That was before, was that before or just after the, um, uh, yeah, for fun and profit.
19:20Izar Tarandach Paper? Oh, that was, that was 1996 paper.
19:27Chris Romeo 6 or 7. Yeah. Yeah.
19:29Matt Coles So you were just before LF-1s. We should, we should recognize, like, like, you know, like Chris said, security has existed for some period of time. It was often associated with more physical and personal security and computer systems. I mean, WarGames was a, classic movie about security, but they didn't talk about security.
19:51Izar Tarandach Security became a thing when The Cuckoo's Egg became an actual pocketbook.
19:57Matt Coles Yep.
19:58Chris Romeo Written by Cliff Stoll, which is awesome.
20:00Matt Coles Listening at home, if you have not read it and you can't have an AI summarize it, you have to read it.
20:05Izar Tarandach And, and if you haven't read it and you call yourself a security person, shame on you.
20:11Chris Romeo Why? I thought we weren't supposed to book shame people.
20:15Izar Tarandach No, we get to throw the book at them, right?
20:18Matt Coles Yeah, we should. We should. It's a small book, so it won't hurt a lot, but we should definitely do it.
20:23Izar Tarandach No, no, seriously, it was like, it was a pocketbook. And for those who don't know the story, Christophe was an astronomer in a university. And he noticed back on those times you would pay, wow, another thing that came around.
20:38Matt Coles Yeah.
20:38Izar Tarandach You would pay for time of use.
20:40Matt Coles Utility computing.
20:41Izar Tarandach On a computer, right?
20:42Matt Coles Or utility. Payment. Yeah.
20:44Izar Tarandach Departments would get printouts with the, uh, the use that they had of computer time. And he noticed a discrepancy of $0.002, I believe it was. So that's like 0.2 cents or something like that. It was, yeah, it was a fraction of a penny, something of a ridiculous magnitude. And he started following it and he ended up, uh, uncovering a KGB operation to, uh, they were targeting, I believe, space shuttle data.
21:19Matt Coles And he had to chase back through the network and it was telephone network. It was telephone, it was modems.
21:23Izar Tarandach It was modems.
21:24Chris Romeo Yeah.
21:25Izar Tarandach Did it ever become a, a movie or am I tripping?
21:29Matt Coles I, I think there's, there's some, there's so many movies that were sort of based on the idea.
21:34Izar Tarandach Like there's a lot of tropes that came from there.
21:36Matt Coles Like Swordfish, I think was a classic one that I think that's the, the best bit of the book.
21:39Izar Tarandach And yes, I'm gonna give it up because I know that you people are just not going to go and read it just because I said. So he had this piece of like fake data and he had to keep the attacker assessing for as long as needed to trace back the call. So he jingled his keys on the back of the modem. And these modems, you people have to remember, they had like this big connectors, like serial connectors, every Every bit was one wire and he jingled his keys over there to put, uh, um, uh, actual, uh, noise in on the wire so that, uh, the thing would degrade but not go down.
22:22Matt Coles And, and that was the beginning of, that was the concept eventually became the honeypot.
22:26Izar Tarandach Yeah. Yep. So many things came from that book. And I think that's that one and the takedown, which is a bit more, uh, uh, of, uh, that, that's a bit more of a Hollywood-y story. That's Shimomura versus, uh, Mitnick, but still, uh, worth, worth reading. Those two, I think, were the things that put a lot of people on the way to what we do.
22:54Chris Romeo Yeah, I don't think the, the one about Mitnick was Holly— was Hollywoodized. No, that one became a movie because, uh, oh, I'm saying, okay, you're saying it became a movie. Okay. Yeah. No, I'm saying the story, like the story was what happened to him.
23:06Izar Tarandach No, no, no, no. The, the, the story as told in the book is a bit already a bit Hollywood-ized. Oh, you mean, you mean like made bigger than, uh, some, some stuff, I believe.
23:17Matt Coles I mean, a lot of hack, a lot of hacking has to be, cuz otherwise it's not interesting, right? I mean, true. It, if it takes, if it takes a hacker, if it takes somebody who is gonna break into a system days to research, uh, I guess that's one thing that, uh, WarGames actually did right early on, right, was it takes days, potentially weeks to research and to pour through old manuals and right.
23:42Izar Tarandach And not like, I'm in, which is actually something that you see in Hackers, the movie, right? When they are dumpster, uh, dumping, dumpster diving, looking for dumpster diving and, and all, which you do not see in Sneakers. But, uh, I don't even start throwing.
24:00Chris Romeo No, no, no, no, no.
24:03Matt Coles And, and actually, so here's an interesting question. Did, are, did those movies influence security people or did they, were people already in security and they saw those movies and they thought, this is how I want to be. This is like, I want to do that. I think for me, those, some of those movies like Sneakers especially influenced me. In terms of like, yes, that's a little bit of an idealized view of security obviously, but that lifestyle, not the lifestyle, the sort of the environment, like it's a problem to solve and you have, you know, that, that it's a specialized environment too, right? This, it was a niche at one time.
24:42Chris Romeo Wait, it just hit me. Did they have quantum in the box?
24:47Matt Coles No, I don't think so.
24:48Chris Romeo Was that what it was? I think there's, I think there's, I think time travel happened and that is actually a quantum computer in the box that's able to decrypt everything. I just realized, solved it.
24:58Matt Coles That might be— it might be somebody with a paper. I mean, well, we don't know the under— you know, it's interesting, uh, Neil Stephenson does a lot of like real research for his books, and, and he did some, you know, Cryptonomicon is the, the sort of the classic there, right? And there's research that's been going on for decades that are finally coming to real life in the computing space and in security specifically. So it's Maybe that chip, uh, you know, from Sneakers was quantum-based. It would make sense.
25:32Chris Romeo You heard it here first, folks. This is breaking news.
25:35Matt Coles And if anyone knows about the script, the script writing and the basis for it of Sneakers, that would be cool to hear.
25:43Izar Tarandach I think it's more like the Clipper chip. Yeah, but the Clipper chip— that was the level of technology that was going on. But back on the influence thing, like if you look behind me, for folks just hearing this.
25:57Chris Romeo Yeah, I see a picture of the Muppets.
25:59Izar Tarandach I got the— no, there's a reason why I have 3 pictures behind me. One of them is Matt's and mine book, because, hey, it's a great book. Everybody should read it. And the other one is the poster from the WarGames movie, because I consider that my, uh, origin story, to tell you the truth. That, that's when I looked at something and I said, hmm, I want a computer in my room. And it wasn't because of, uh, who was it? Elizabeth Chu? But, uh, yeah, it may have contributed. And of course the Muppets because I have to say something about community, right? So, but I think that Matt again put his finger on it with the, uh, the, I think that those movies, they, they helped, they, they helped. It wasn't a red button this time. They, they help build the, uh, the romantic version of what we do.
26:54Chris Romeo Right.
26:55Izar Tarandach And a lot of people got onto that bus because the romantic version of it was so compelling. But, uh, as most everything in life, once the romantic version meets the real life version, some people notice that, uh, that's not what I signed for. And some people get it more. Even stronger, hey, that's what I want to do with my life. And I think that most of the people that I have met that were in some way in security decided to, to, to stay. So to me, what that shows is that in a way or another, it is a very close-knit community of people. We do have more similarities than we have differences.
27:40Matt Coles Well, and, and actually that it's interesting that you say that, um, you know, a while back, probably 10, 15 years ago now, the community of people who did product security was a lot smaller. Oh yeah. Right. Then, then IT security, right? So security had different flavors or still does, but you know, there was, there were these niche groups and product security was a, I mean, at one point we were we, some of us were thinking, well, how many people are actually doing product security? Because at that time, I mean, Microsoft obviously really kicked it off with, with Bill Gates, not kicked off, but made it, made it public, right? There was, there was stuff happening, but, but that really triggered it. Yeah. And then, but not all companies were doing product security, at least not making it well known, or they had a handful of people in that environment. So when you went to Black Hat or DEF CON or one of those security conferences, I mean, BSides, I think. Sort of appeared as a result of some of this niche, these niche conversations. And at some point when we turned to threat modeling, right, threat modeling to the collection of people who do threat modeling was a, was a niche, you know, subgroup of people. Not everyone does it. Now everyone's doing it.
28:58Chris Romeo Right.
28:59Matt Coles Or should, or should. Well, and you know, and there's maybe, they may be doing it sort of begrudgingly or it's sort of just now happens. right, as part of their regular work. But, you know, the people who were heavily involved with making threat modeling happening was a small group of people. Does that make it any more or less exciting now that it's starting to become ubiquitous? The things that we do, the things that were niche, was the niche the important part?
29:35Izar Tarandach I'm gonna confess something here.
29:39Matt Coles You're not really a czar?
29:41Izar Tarandach I'm a pod, I'm a pod. But the confession is, I think that by now it's very clear that I have a certain passion for track modeling.
29:53Matt Coles Gee, you don't say.
29:55Izar Tarandach But lately, it's become annoying. It's become annoying because The amount of people insisting in looking for the silver bullet, the shortcut, the accelerator, the thing that's going to do it for all of us and going to do it perfect. People, just fucking do the work. It's not that difficult. Sit down, look at the stuff that you're building. Think about what could go wrong. Think about what you're going to do about it. Think about how important that is for you. And then see if you did a good job. It's really not magic.
30:42Chris Romeo Okay. We have officially reached curmudgeon level 10.
30:46Izar Tarandach No, no, no, no, but, but really, he's got a point.
30:49Matt Coles No, no, developer developers. Think about the APIs you're calling and use the language like you're supposed to. It's not magic.
30:56Chris Romeo I just have a feeling he's gonna yell, get off my lawn at the end of this.
30:59Izar Tarandach Chris, Chris, no, no, no. My point is, my point is, please come into my lawn and stop creating those little patches of grass that all of a sudden are going to make the whole thing shinier and, and easier and whatnot. People, it's very simple. There is a problem. It needs a solution. A lot of it is conceptual. What you think could, you could plug into the conceptual thing right now. We have, I'm loath to use the word proven, but has been shown to be a glorified checklist. So get off that bus and just freaking do the work. It's not going to help. It's not going to help the amount of emojis that you put into it. It's not going to help the amount of abuse cases that, like, what you're doing is an abuse case.
31:58Chris Romeo So your frustration stems from people that are entering threat modeling and trying to, they think they're coming up with new ideas or something that are going to revolutionize or make things move so much faster. And you've been around a long time. And so you know that that's not going to be how it works. Is that?
32:17Matt Coles What I'm hearing?
32:18Izar Tarandach We have tried every single accelerator out there. Outside from creating something that can actually do the conceptual work in a blaze of speed, you are not solving the garbage in, garbage out problem. Until you solve that, you can be as smart as you want. You got horrible information coming in. You're going to get horrible information going out.
32:48Matt Coles Haven't you met Claude? Have you met Claude?
32:51Izar Tarandach Yes, but have you met the people who feed Claude with information? Because just saying analyze this codebase, I'm sorry, we all agree. That's not exactly trustworthy.
33:03Matt Coles So this is what goes around comes around, right? Or what's old is new again. Remember the early days of static code analysis tools?
33:13Izar Tarandach I remember.
33:15Matt Coles And the garbage you would get from the garbage you would put in.
33:20Izar Tarandach No, but that was in part because basically we were so stuck in regular expressions. Remember when ASTs came out, what a revolution it was. All of a sudden, instead of 100,000 false positives, you had only 10,000. Life became so much more, so much less unmanageable. But the, the, the point that I'm trying to make is the problem of threat modeling has been solved. We have methodologies, we have proven results. We know the ABC that leads to Z, right? What hasn't been solved and what I urge people out there to please go and do is focus on the problems around threat modeling. The other things that make threat modeling possible. Got some startups working in that space, some beautiful work, but I still believe that there's a lot of stuff to do in there and that's the specifically interesting place to look at. And the way that that joins the personalities is that once upon a time, we were able to look at the problem, take a step back, see what's around the problem, and solve one thing after another. Nowadays, because of the short loop of let's make money very, very fast, everybody's solving this problem that all of a sudden we have this very shiny tool that's all apparently solves. So we all got the generative hammer and every problem became the prompting nail. It's not.
35:11Matt Coles At some point we should talk about whether AI is a member of the security community with a personality. It doesn't have a personality.
35:20Izar Tarandach I'm not saying it is, but.
35:21Chris Romeo Yeah, let's talk about that.
35:21Izar Tarandach No, it isn't. Next.
35:23Chris Romeo That was quick.
35:26Matt Coles But it fits that role. It fits that role, right? It is your developer.
35:29Izar Tarandach No, people want it to fit that role.
35:30Matt Coles I know. It's, it's your, it's, but it's your developer today, right? People are not looking at it as a tool for accomplishing a goal. They're looking at it as I have a friend.
35:39Izar Tarandach Because it is today's silver bullet. People, it doesn't matter how many times it tells you, oh, you are right. Yeah. I made a mistake. It's not your friend.
35:51Chris Romeo Wait, what are you trying to say?
35:55Matt Coles Oh, sorry.
35:56Izar Tarandach I was wrong.
35:56Matt Coles Don't listen to that, Claude. Don't listen to that.
35:59Chris Romeo It's my friend. All right, well, we're outta time for today. We will potentially revisit that episode, that idea of does AI have a personality, if any of us can remember it a week from now.
36:13Izar Tarandach Oh wait, let me write a markdown file for us to remember later.
36:16Chris Romeo Yeah, tell Claude to remind you.
36:18Matt Coles I was going to say, can't you ask Claude to do that?
36:20Chris Romeo Yeah, just tell Claude to remind you, and then 2 years from now, you'll get a reminder. I didn't know that you wanted it in a week. Yeah, but friend.
36:27Izar Tarandach Wait, wait, wait, wait, wait, wait, wait. So, uh, again, we are approaching, uh, OWASP Vienna at the end of the month. Uh, 2/3 of us are going to be there. So if you're attending, if you're listening to this, And if you want to go on our faces and say, you're wrong, you're wrong, you're wrong, you're wrong. And here is to say, you're right, we're wrong. Uh, then catch us there. We'll have nice stickers. First time that the podcast is going to have stickers.
36:59Chris Romeo Uh, Matt says, hi, my name is Slim Shady.
37:02Izar Tarandach I'm always going to go with Nino Montoya, but now it's very common to see people with 6 fingers, so. It's not a big thing anymore.
37:11Chris Romeo Yeah, it's true. All right, well, we got to wrap this episode up. Thanks for listening to Security Table, the original, original OG, the one.
Transcript supplied by AssemblyAI. It may contain transcription errors.
Originally published as The Security Table. Part of the AI Security Table archive.
