Musings from the Cyber Trench
Musings from the Cyber Trench
Energy Is the Upstream Cyber Risk in Power Infrastructure | Bethun Bhowmik | EP 112
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
Energy is the upstream of everything
In this episode we explore how energy infrastructure has become one of the most critical and vulnerable systems in the modern world
From power grid attacks to systemic risks this conversation reveals why cyber threats to energy impact entire nations
Responsible for ICAM, Zero Trust, or identity security in a federal agency, prime, or large regulated enterprise?
If you’re trying to move from strategy to execution, start with Zephon’s Zero Trust Readiness Assessment: zephon.tech/zt
Questions or guest ideas? Email defend@zephon.tech
Hello everyone. Welcome to another episode of Musings from the eCyber Trench. Today we are going to talk about energy, cyber, and uh and competitiveness. And tied to all that, the risk lurking in our critical infrastructure. Our guest today brings a rare perspective that sits right at the intersection of technology, capital, and critical infrastructure. He is a five-time founder with three successful exits, an operator-turned investor, and someone who has spent more than a decade building and scaling uh multiple businesses across global markets. He started his career as a software developer at Oracle, went on to help scale the uh the Techstars community program across six countries, and later worked with high growth startups across India and Singapore. He moved to the US to deepen his focus on finance and data, earning an MBA in corporate finance and a master's degree in uh in data science. Where he works closely with US utility companies, helping power more than 120 EV infrastructure projects. And he is now building the first AI native infrastructure fund to focus on the future of energy and compute. Interesting times for sure. And here's a fun fact. He also happens to have represented India as a national level computer student. With that, please meet my guest today, Beethun Gomek. Welcome to the musings from December Trans Beethun.
SPEAKER_02Thank you so much, Vishwal. How are you doing?
SPEAKER_03I'm doing good. Uh trying to stay warm. And today is January 4th at the time of this recording. I'm in Dallas, and like most of the country, we have snow, ice, and rain and sleet. And uh lot of folks are under power. So our uh uh have power cuts right now or will lose power in the coming day or two. So our conversation uh today is going to be very apt based on where we are, especially today.
SPEAKER_02Oh, absolutely. Uh first of all, I hope you don't lose power. You and your family stay warm. But uh yeah, since I work with so many utility companies, my email have been blowing up from various utility providers giving constant updates regarding the the storm and the sleed situation. So I think uh overall we should not be worried, but uh yeah, I guess everything uh just uh is is nice over the next couple of days.
SPEAKER_03Hope so too. I know we can't really control the weather, especially when it comes to ice, but let's talk about what we can control. So if a nation wanted to uh undermine another country's competitiveness without firing a single shot, is critical energy infrastructure the most effective place to do it? And why?
SPEAKER_02Sure, I mean uh you can you can say that. Uh and let's think of it this way energy is the upstream of everything, whether you're talking about running a factory or or ports or hospitals, even uh the the the cloud uh and AWS uh data centers, uh, which is by which we are probably doing this podcast, everything is dependent on energy. And uh I think two interesting uh examples come to my mind uh where uh this may be apt. The first is probably I think uh Ukraine 2015. That as far as I know was one of the first uh attacks by a cyber uh firm on a power grid. So what they did was potentially it was a Russian-linked uh uh cyber attackers, they disrupted the electricity for over 2,000, 200,000 customers in Ukraine. And they put some kind of malware uh to disable the substations and delete the files, and it took some time for operations to restore power. So that is that is one classic use case. And then the other thing which is happened in the US in 2021 was where there was a ransomware attack by a group called Dark Side on uh the colonial pipeline. And this happened around 2021 and it forced a six-day shutdown. Uh yeah, I'm I'm sure you would have remembered that. Uh lots and lots of people were affected uh because I think uh 5,500 miles of pipelines were halted, and uh they actually the colonial pipeline had to actually pay 4.4 million dollars as a ransom. At that time it was 75 bitcoins, I'm sure if you convert the Bitcoin price right now, it's yeah. Uh but uh that is another classic use case where uh like the cyber attacks can literally bring a country or stop like date-hood operations of a country or or region of it for sure.
SPEAKER_03No, I'm just uh one side, like a bit of a tangent. I I should have asked my clients to start paying me uh in Bitcoin a long time ago.
SPEAKER_02Oh yeah, yeah, absolutely.
SPEAKER_03You can always you can start now, Bitcoin is down a bit, so you can uh when we talk about energy production and uh and and distribution, what parts of the system are most misunderstood by policymakers and uh and uh business leaders from a cyber risk perspective?
SPEAKER_02I think there's uh three misconceptions, I would say. The first would be that the grid, as we talk about it, is just one system, which it is not. The grid is literally a patchwork of different kinds of uh uh different owners, different vendors, different contractors, different legacy control environments. So it's like a lot of dots that makes up the grid. So it's like, and all of them are working in harmony, which is why it's great. But uh all of them have to work properly for the grid to function properly. I think that's the first misconception what uh a lot of people not uh uh who do not may not have the technical uh uh knowledge may not understand. The second misconception would that the grid is uh it's it's not uh and from a cyber risk risk perspective, it's it's different because it's both a physical layer compared to a software layer. So when in the in there's a cyber attack on a on a software perspective, it's like we have to create some safety patch and make sure the uptime and those things are addressed. Out here, there's actual uh issues because there's a lot of back-end hardware operation technologies involved. And the third part is that those the hardware backend technologies have constraints. For example, like if a transformer blows off uh or there's a critical equipment failure, a utility company would have to mobilize crews. Now, crew mobilization can take a week or two weeks, and then we have to go, and this is assuming that the utility company would have that reserve transformer in stock. If that is not there, then procuring those is like another additional lead times. So that is what is very interesting. That most people uh may think that it's uh when you talk about energy or electricity, it's like a highway, if I'm taking highway analogy. But just because if I have a highway doesn't mean that you can reach your home. To reach your home, you have to have an off-ramp or an exit, and you can use that exit to reach reach your home or wherever you want to go. So if that exits, or in this case the substations, which I'm calling as the exits, are compromised, then that uh triggers the anti-system failure.
SPEAKER_03You mentioned the grid to be a patchwork, you know, and there's that's very rightly said, you know. So based on what my limited knowledge is of our uh national grid, we our systems are old and uh and and uh they are aging. There are a lot of things which are still done manually, and uh we have uh and we also have uh fragmented uh ownership all over the place. Does that cause uh inefficiencies when it comes to being able to defend us from a cyber attack? And to what extent?
SPEAKER_02Sure, we can say that. I mean inefficiencies, yes, definitely. It does cause inefficiencies, and let's look at it logically. If there's inefficiencies, those inefficiencies create workarounds. And uh workarounds create uh access paths that can be uh utilized by cyber attackers, so it's all like a chain uh of one where one leads to the other. And uh let's just I mean, to break it down, let's just take an example, right? I mean, let's say uh the the utility company creates an emergency remote access for a contractor to uh for uh to do to do commissioning and to like to share what commissioning is with the with your viewers, it's basically testing. Like there has to be pre-energization commission commissioning and post-energization commissioning, which means before any site turns on permanent power, every equipment, everything has to be double-checked, whether it meets the standards, whether there's any fault current issues. There's a lot of studies that happens, like the arc flash study. So to make sure nothing breaks, the equipment doesn't break uh in case of uh any outage uh or like power surge. Uh so let's suppose I create uh or the utility company creates a remote emergency access for a contractor to do some work just for a week. And then they forget to remove that. So that temporary door would tends to stay open, and those can be easily be leveraged by any cyber attacker to enter the enter and disrupt your uh systems and processes. So so that is definitely there. Uh so there's a lot of uh uh the the the patchwork, the fragmented ownership just leads to more uh inefficiencies and workarounds, which uh I uh as long as those back doors are closed and somebody's keeping uh a watch over that, there's there's no uh uh system-wide uh uh uh weaknesses that can be exploited, things should be good. But the realities may not be that. And as long as those temporary walkarounds create different potential access paths, there's always a cyber risk that can be uh exploited.
SPEAKER_03Okay. You think we are in need of a of uh of a compliance framework to make sure that we are moving in the right direction?
SPEAKER_02Oh, absolutely. I mean, uh I'm sure there are there are compliance frameworks. It's uh it's basically uh are how efficiently is that compliance framework? Are we just taking off check take tick like checkboxes on or are we actually uh making sure that we are proactively removing potential risks? Because I'm uh I'm as I work with a lot of utility companies and they're they're just uh and commissioning, and there's definitely a lot of compliance checklists with without which the the city or the county would not give you the permits. It's called authority having jurisdiction under which jurisdiction you are. So it's uh checklists, it's just those are legacy checklists. Those checklists may not uh have use cases that from a cyber angle or from a modern AIA era where uh may not have captured a lot of the issues. Classic, classic example, like one of the utilities where I was my team was uh filing an application for, it's a small cooperative utility. That utility in the application form had an option for us to provide the drawings, uh electrical and civil drawings, via a disk. So a CD disk. So they still have that option of uh I've seen that too.
SPEAKER_03Yeah.
SPEAKER_02So the fact that uh which means that their the form has not been updated in some time and there's legacy systems and stuff. So it's not about not having compliance. I'm sure compliance checklists are there. Are those updated with the rapidly changing environment and we are living in?
SPEAKER_03I once had to submit one uh one proposal for uh a Texas state agency where I had to submit it in a PDF form online, also copy it to a USB drive and print it out and ship the USB drive with the actual physical paper by mail. Yeah, interesting. And correct me if I'm wrong here, the US has uh East Coast grid, a West Coast grid, and then Texas has its own grid. So you think it's good to still keep them divided, or it's good to have one national grid, or we should or or is it better to like fragment them into multiple smaller grids so we can uh we can uh reduce the potential uh impact of a cyber attack?
SPEAKER_02I mean, it's not a question of what I think. I think it is what it is. There's certain advantages and disadvantages for sure. For example, uh our reason a lot of the Texas is becoming a hub is because Arcot, the the organization that manages the Texas grid, is uh not within the purview of the federal uh the FERC, the Federal Energy Regulator Regulatory Commission. And so that's good because the average uh time to get power for a site, if it's a an uh like let's say a data center, right now can range between two to four years. But for Texas, yeah, that is uh something I delve deep into one in my book, one of the books a book that I'm writing. But there's basically three clocks to power. And let me delve deeper into it. Uh, this has led a lot of loss of capital for a lot of people, where people have just tried to build and the building uh and developers uh uh and assuming that they would get power, but then what has happened is it's like the same thing. Uh like uh just because there's a major power line passing uh by the by your house doesn't mean that your house has the power. It's like the same thing as the highway can pass by your house, but if there's the access road to the house, you don't have power. So average construction timeline can range between 12 to 18 months. But if you're if you don't get power in three years, that means you are sitting on dead capital. That means you, as a developer, are taking on a hundred million dollar project with bank loan, some of your own capital. Uh you have completed your project in 18 months and then sitting for 18 months more just because you don't have power. So that is the critical issue uh we are being faced right now. And it is truly a once-in-a-lifetime kind of a scenario we are being that we are facing right now. And the reason for that is the grid is being hit by two sides. I mean, it's uh the data center explosion, the especially the AI data center explosion, and the electric vehicle uh charging explosion, right? 2024 the the peak load capacity for US was around, I think, 33 gigawatts or something. By 2030, it is expected to be 166 gigawatts. So that is like a five times increase in just six years, which is never in the history of uh uh the modern world hazard happened. And let me let me uh contextualize that would mean it's like adding like 12 uh New York Cities peak load in just five years. So imagine building the power 12 New York Cities from magic from thin air. So that is the kind of problem we are being faced, and historically, grids have seen the power demand rise linearly. For the first time, they are seeing a hockey curve where it's like going up crazy, and they even like that's the that's the challenge that's facing us for us to leverage all the Gen AI, the trillion dollar gains, that can only happen if the backend infrastructure is there to support it. So we are truly living in one of a uh kind uh scenario. And so that's why it's very important that we understand that uh uh there's three clocks. One is the the time to power, the time to permit, and the time to, of course, build. Uh, if any of these clocks is are not synchrono synchronized, it can lead to massive dead capital or dead spending. So just to wrap up your point, one of the reasons I think Urquot, you said whether it should be separate or thing. I think one of the things Texas is having uh is becoming a hub for a tier two level uh data centers is because in Texas it's faster, you can get power a bit faster. The average uh timeline to get power is around between 12 to 18 months compared to the national average. So a lot of the hyperscalers, whether it be Amazon, Google, Microsoft, are exploring the the Dallas Fourth Worth, San Antonio, uh, and the Austin belt. Uh, this is one of the biggest reasons because if you don't have that uh that uh shorter timeline to power, you're losing out a lot.
SPEAKER_03I was not aware that it's like so so bad in the rest of the country and so good in Texas, you know. I thought they were just coming here because land was cheap, you know.
SPEAKER_02That too.
SPEAKER_03Yeah. So you mentioned like uh the uh the uh hockey stick curve. What uh percentage of it, if you know, is attributed to uh to AI and what is attributed to the uh electric vehicles?
SPEAKER_02Well, I told you, right? The power, the peak load forecast would is expected to go up by five times in the next until by 2030. So if I have to look up the breakup, what is driving that? 55% is data centers, AI data centers.
SPEAKER_03Uh 5% you said?
SPEAKER_0255.
SPEAKER_0355. Oh wow, okay, yeah.
SPEAKER_02The AI data centers is driving majorly the growth. So 55 is expected to come from AI data centers. Industrial manufacturing would account for around 30%. And then the EV or the electric vehicle charging infrastructure would account for 15%.
SPEAKER_03Okay, so that's still comparatively low to the others.
SPEAKER_02Yeah, yeah. But if you add these up, they just basically, yeah, that's I think that's the 100% breakup. But these are the three major buckets, uh, what is driving the crazy growth.
SPEAKER_03You think it might, and this is just a thought that came to my mind that we should pause on the electric vehicle adoption so that we have more energy to and shift it towards like more energy uh portion can be shifted towards uh. AI and uh and uh racial production. It's a thought. Yeah.
SPEAKER_02I mean it's a good thought, but I mean uh the big question is will companies do it because eventually you have to think of it from a PL perspective, the what uh improves the bottom line. Now, uh again, if you look at the Department of Energy's uh report, currently there's five million EVs on the US roads. That number is expected to grow to 33 million by 2030, which means actually a 7x jump. And the jump in that seven times jump would not come from people like you and me driving our electric cars, it would be driven by uh more on the commercial front, where uh Amazon, FedEx, uh and uh Walmart are uh reverting to would revert to electric trucks to deliver their uh to the the connect the middle mile to the last mile. And so the reason they would do that or they are doing that is because it is uh I mean it improves their bottom line over over, of course, a certain number of periods they have to make investments and they have already made those investments, so there's already a lot of uh game that is in play. So I don't uh I don't I can't comment on how the the the personal EV uh market would uh would grow in the next five years. Uh but from a commercial perspective, I definitely see a lot of growth happening uh with most uh companies switching to EV trucks in instead of ice ice vehicles.
SPEAKER_03I think uh the self-driving trucks which are being uh being promoted, they also would uh most surely be uh be electric.
SPEAKER_02Yeah, for sure.
SPEAKER_03You mentioned your book, and if I'm not wrong, it talks about uh data center arms race. Can you tell us a bit more about it?
SPEAKER_02Yeah, so the book that I'm writing is it's the name is the Power Investor, and it's basically how anyone could uh uh capitalize on this once-in-a-lifetime opportunity to uh and use strategic leverage to to close deals worth hundreds of millions. Uh but uh the premise of that is that yes, we are it's a race happening, and it's it's not a race for real estate or land, it's a race for constraints. Now, what are the constraints? The there are three constraints which are also the three choke points for any data center. The first is the power, which is basically a substation. Does is there enough power to serve to or to service the data center? That's the first key question. Uh uh, if not, the data center does not get built, as simple as that. Or they would have to, you uh anyone who's doing that would have to uh think of a lower load and build a data center in phases. The second constraint is the latency, or how quickly does data travel from the data center to us when we're doing the podcast or or or to Chat GPT's servers when they are basically training their next uh model. Now, uh, if the latency is not there, then the data center is just a warehouse. The the the what makes a data center is a data center is that there's fibers uh which uh which promise immediate speed for the data to travel. Uh and then the third part, the third constraint is the water rights or the cooling. Like cooling is a major issue. And if that is not if that is too costly or if there's constraints from the local county, uh as it is in Arizona, where where where water is expensive, then that creates an issue. So that's that is what I have uh based on the experience of talking to lots of experts, uh, and just understanding the industry from the inside, uh, is what I call the three choke points. And yeah, I delve more into that in my book. Understood.
SPEAKER_03Are all three choke points uh uh do all three choke points carry equal weightage in your mind?
SPEAKER_02Not equal, but like you cannot do without either. Let's just say they're non-negotiables. You need all three. Uh and and uh technically to your to your since it's a cyber security podcast, each of these things can be has a cyber risk. Let's think of it this way. Like uh I've if someone was to make us had do a cyber attack, there's an inst internal constraints and there's external constraints. The three choke points that I told you, which is power, latency, and water rights, let's break them down. So if if someone attacks a substation and takes out critical equipment, the power is down, data center is out. So clear use case. Secondly, if the most data centers have different kinds of service providers, multiple service providers like ATT, Verizon, Lumen, uh for for the data, uh data fiber for the latency part. If I compromise, if I'm compromised on any one of those, if ATT network is compromised, that hampers my ability to uh get data real time and a cascading effect. The third effect is the cooling, and so there, of course, there's uh for cooling water is required. And then if if you hit critical water infrastructure from where the data center's plumbing is happening, then again, that again shuts down the data center because otherwise the equipment would get heated too much or the server racks. So that's the internal, those are the internal constraints, those are the external constraints, my bad. Internally, let's understand again how cyber threats can happen. So data centers is not a new concept, right? It's just the AI data centers is a new concept. We still use data centers to host websites and other uh store files on uh Dropbox or something. Now, those kind of storage would have taken like uh uh uh five kilowatt uh or or let's say whatever it is takes. The AI data center's consumption is 10x to 20x that because it takes more power to train a model, to train a model and stuff. So you're talking about higher density, where uh and that compute capacity is what we talked about, that keeps changing every four years. So uh let's say somebody tries to buy an old data center and refurbish into an AI data center. Now, what would happen is that there's a lot of uh, firstly, they would need to come improve the compute capacity, and then to do that, there's a lot of uh up during the upgrades, there's a lot of cases where cyber attacks could be possible because the developer may not know the critical points of uh of how some uh loopholes can be leveraged. Uh so that is one. The other thing would be the PUE, which is called the power usage efficiency. Like uh, and that again is like comes to the cooling part. So in an ideal case scenario, the P the power usage efficiency should be just one, which means if 100% of my power or electricity is going towards the server racks. Theoretically, that's never the case. Most data centers have like a PUE of 1.3 to 1.5, which means that if I'm saying a PUE of 1.5, which means 100% is going towards the powering the chips and 50% is going towards cooling to ensure that the data the racks are in the proper temperature. So uh there again, if you're buying an old, let's say 2018 building or a data center, those there, the PUE could be 1.8 or 2, which means 100%, like if 100% of your power is going uh like going to the chips, you're spending 100% more just to cool the cool the racks or cool the servers. So these are all uh areas where uh vulnerabilities exist. If some uh cyber group is creative enough and have an access point, it can potentially be leveraged.
SPEAKER_03So uh uh legacy data centers can potentially have a much higher cyber risk compared to the modern ones.
SPEAKER_02Absolutely, by design. So if somebody's doing that, there has to be again a lot of checks and balances uh for that.
SPEAKER_03If you had to give energy operators and infrastructure leaders three priority moves to reduce cyber and national security risk, what would that be?
SPEAKER_02Three moves for reducing the cyber risk. Yes. Yep. Keep the systems re look relook at the existing systems and update those, have like accountability just because every three to four years the entire game changes in this uh industry. So make sure the systems are being updated uh and the processes are updated. Remove any kind of, or or maybe it would be good to understand where there's potential uh access that can be leveraged, or uh uh so maybe doing uh flagging those timely. Uh a lot of the times where some construction happened, they would have given some temporary access that could be flagged maybe a few years ago, just making sure that timely those are removed. And then uh maybe running some restoration drills under degraded conditions, like the the what's happening right now uh with the with the freeze uh right now. So that that means that uh uh most utility companies, I'm assuming, have assumed a worst-case scenario to uh uh uh where some of the transformers or critical equipment may get disrupted and they would have those equipment in stock, their crews mobilized, uh uh to make sure that if something goes wrong, uh uh those can be like the power can be uh restored at the earliest. So you have to always same in the in the military, like around the worst case scenario, drill drills and doing those if you don't would help you be more prepared. Because if you don't have, let's say I need a 2000 KVA transformer, I don't have it. Uh so procuring that is like the lead time is 52 weeks. So that's uh definitely a failure situation.
SPEAKER_03So just running those. You mentioned uh procurement a few times, you know, like is supply chain, you know, from the cybersecurity point of view, it mainly pertains to software. Right, but based on what you're saying, you know, in critical infrastructure, especially when it comes to energy production, it's a lot more than just software. Like there are so many physical components to it, and like the supply chain risk, even sourcing, uh manufacturing, all that come into play.
SPEAKER_02Yeah, and that uh that sourcing lead time has gone up by three times since we are in this hockey stick hockey stick curve. All the if you're looking at transformers, switchboards, switch gears, uh relays, everything has the the time to procure has gone up by two to three times, which is which is a bottleneck. So yeah. And I'm I'm seeing it day in day out in terms of for critical projects being delayed because the transformer can't be procured on time.
SPEAKER_03And this delay is because we can't because we're just producing slow, or the consumption is so high that it cannot meet the demand.
SPEAKER_02The other one, the second one.
SPEAKER_03Okay.
SPEAKER_02Overwhelming consumption.
SPEAKER_03In terms of like uh procurement, like uh or uh uh or manufacturing, are most of these being manufactured in the US or is from China or all over the world?
SPEAKER_02That's a good question. I would not be educated enough to comment, but depends on uh the it depends, like if it's a utility, I would assume, and my assumption it's it's locally procured, local supply chain. No, it's some it's some if if it's someone like Amazon uh or or uh or uh Microsoft who have a global supply chain, they would definitely try to go uh and create a globalized global supply chain leveraging economies of scale, economies of scope. So it's uh depends which lens we are looking at at that from the supply chain.
SPEAKER_03Uh you mentioned data centers uh and the amount of uh energy that they used. Uh I was at the EarthX conference last year and there was a lot of talk about using uh mini or I think micro they call it, micro nuclear power plants, like which would feed these uh recenters. In fact, if I'm not wrong, uh Microsoft is trying to work with Constellation uh Energy to I think bring an old uh nuclear power plant back to life. I forgot where it was. And then Google and uh is also trying to work with a company which is creating uh mini nuclear power plants. Uh Amazon is also has put an uh ownership into a company which is which is doing that. So you see uh like potentially us going more into that direction?
SPEAKER_02100%. I mean uh that is the way you have to go because uh the grid is just the grid. Now, when we talk about the grid, we just assume 100% of that is being powered by electricity. But now, in the in the recent future, that has never been the case. You'll see now you'll see grids where 80% is being powered by electricity and then 20% from solar or wind and other renewables. Um more and more we will move towards other renewables or hybrids, where it is either powered by SMRs, which are modular nuclear reactors, as and then solar, wind, geothermal, or a combination of all of these things. Now, electricity production is unlimited. So, very like we are reaching a will or we will reach a world very soon where most grids would be hybrids and not just electricity alone. And that's where the microgrid uh phenomena will come into play. And it is already in play, it's just it will become more mainstream.
SPEAKER_03What is one uncomfortable truth about energy and cyber resilience that uh leaders in the critical infrastructure uh industry and energy production need to accept?
SPEAKER_02That is the that is the premise of my book, which I call the the three clocks problem, where leaders uh in the infrastructure space has to understand that how do they synchronize all the three clocks, and then there's a fine dance between the power clock, the permit clock, and the construction clock. If that uh if those are not aligned, then it'll it'll just lead to like uh dead capital and uh lack like wastage of resources. So so that's that's a brutal reality. A lot of funds have not realized, and a lot of people have lost money on, where they have just built powered shells, which are like massive buildings, and it's a uh it's just a shell because there's nothing because there's nothing can be done because the other two clocks are not aligned. Uh whether it's the power clock or uh there's when I talk about the second clock, which is the permit clock, a lot depends on that. For example, if a data center or even if uh any any kind of like an electric vehicle charging station is in a floodplain, that's game over. You'll never get the permitting uh if if it's in a hundred-year floodplain. It's uh there's a lot of uh issues that come with it. Again, if it's a if there's if there's a chemical factory or some kind of soil testing that that that's uh that's uh there, again, it's like you are stuck. That process will be stuck on forever. You cannot have uh a data center near an airport because if it's somewhere near a flight path, the more than the noise, the particulates from the air can actually affect the servers. I know that.
SPEAKER_03Okay.
SPEAKER_02Wow. Yeah, yeah. So like there's a there's a list of checklists uh which uh has to be adhered to uh for even selecting a data center site. Uh if the shape is irregular, that can be a hazard. If the slope or the gradient is more than 5%, that can be an issue. If it is very close to a residential place, then that's an issue because the servers create some noise. So that would be a bothering for the residential areas if it's very close to that.
SPEAKER_03Let's shift gears a bit. Let's let us all get to know Bethune a bit more. So you have lived and worked across multiple countries, India, Singapore, and now the US. How has that global perspective shaped the way that you think about risk infrastructure and resilience?
SPEAKER_02Well, I don't know about uh risk infrastructure and resilience, but uh I am a product of uh meeting different amazing people, people such as you, people who have come into my life from various uh avenues. And I definitely think uh traveling to different places, even globally, even within the US, within sing within Singapore, India globally, has I have come into contact with amazing people, and sometimes all it takes is just connection with one good person to change your thought process or idea and just uh change the trajectory of your life. So uh yeah, I mean I'm big time a believer of uh uh trying to understand the mindset that drives the top people, the top one person, the top five person people, and I think uh uh there's uh there's a way that those can be deconstructed. So uh uh whether it's within the world of athletics to entrepreneurship, uh it's always the very similar uh data points. So yeah, I think meeting a lot of people has definitely shaped me in whatever form. I'm still in my infancy stages, still figuring things out. But yeah, I would say uh uh uh that's definitely something good. And of course, uh I love traveling. Uh I'm a big foodie, so traveling to different places, exploring the cuisine, those are the the perks which anybody can enjoy. So I'm in I've enjoyed those as well.
SPEAKER_03I'm in the same world when I'm traveling. I I'm I actually look at uh at places uh to go to eat first before I make other travel plans, you know.
SPEAKER_02100%. You may only go to a restaurant one time in your life, so why not uh like the best restaurant in the area and then go check it out?
SPEAKER_03So uh uh you have all you have been uh founder, operator, and now and uh an investor. Which role taught you the most and why?
SPEAKER_02I think all the roles. It's just uh you're just wearing different hats, but you're always on a learning, learning mode. Uh and uh everybody's I believe everybody's uh life journey is different. Some people find their life journey soon, which also means their purpose. Some people have to explore a lot, but uh yeah, that's why everybody's life journey is unique. And uh for me, I have explored a little a little bit, I would say, but each one of those has made Me or give me the ability to connect different dots. How in different stages, what I have done. I now there's a look back, and it's like Steve Jobs said that, right? You can only connect the dots looking backwards. So as I look back, it's like this all of the dots have connected in who am I today and what I'm hoping to accomplish. So I would say everything, every role that I was in has its own importance.
SPEAKER_03So while you were like uh transitioning from let's like talk about transition then. So you have moved countries, you have switched careers, and you have moved from going from building companies to now backing them. Right. So which transition for you was the hardest?
SPEAKER_02I think the move to US was hardest. Uh not hardest, but it was hard because it was like uh restarting your entire career. It's just like uh rebuilding. I was all already in a mid-career good trajectory path, but hey, that was the that was the fun of it. I mean, it's just uh yeah, I mean, I was grateful that I got a full scholarship for my MBA. Uh I was grateful. I'm grateful that I got into Harvard to do uh my second master's after my MBA. Uh uh there's a lot of uh opportunities out here. So uh uh I wanted to explore the world of strategy consulting, which I was again I'm I was grateful to get into. So uh yeah, sometimes I think uh you just have to uh list like a bungee jump, you just have to jump before you realize uh uh like what's uh the how is how the jump is gonna be. So but in that analogy, yeah, I think that was a bit hard because it was like restarting everything from scratch in a foreign country, uh away from your family. So it's like it's like it was like a reboot.
SPEAKER_03So when you came to the US, where did you land first?
SPEAKER_02Uh Chicago.
SPEAKER_03Chicago? Okay.
SPEAKER_02Notre Dame is just I think uh a couple of hours from Chicago. So Chicago is was my port of landing. Uh I still love deep dish pizza. Uh my kids do too.
SPEAKER_03Yep, yeah.
SPEAKER_02Yeah, yeah. I I I don't know if I can have it more than once a quarter because it's really very heavy.
SPEAKER_03Very heavy, yeah. Just one piece of is enough, yeah.
SPEAKER_02Oh, yeah, yeah, yeah. Uh but Chicago was fun. I was uh so when during my MBA uh admission, uh the interviewer asked me, and this is when I was in in in India during COVID, that hey, hope you're ready for the winters. So my assumption was the Indian winters. When I came to Chicago, I realized it's like everything is uh like minus 10 degrees in Celsius. Uh like then I realized, okay, this is I didn't sign up for this, but okay, of course I was just kidding, but uh it was fun. So there was a different in context when she told me that I uh yeah, I love winters. I'd be happy to come over and enjoy the winters. But then when I came here, I realized, okay, this is a different winter, the Midwest winter.
SPEAKER_03Yeah, I remember. I was in, I lived in uh in Waukegan for I think 13 or 15 months. It's like north of Chicago, uh, near Guurney. It was brutal. It was, yeah, like I like I don't miss the winters for sure.
SPEAKER_02Oh yeah, yeah, yeah. I'm sure Texas has has that advantage as well.
SPEAKER_03Absolutely. So now let's move towards uh uh towards competitive shooting, you know. I'm sure it requires intense focus and also intense uh discipline. Are there lessons from that experience that still influence you on how you make uh decisions?
SPEAKER_02So just to give some context, I think uh I got into shooting because I went to um one of the top military schools in India, and I did want to get into the defense, uh, but I wanted to get into Air Force, uh but um I would have been unfit because of my eyesight. It requires proper six by six vision to become a fighter pilot. Hence I decided to move out. But uh while I was there, I picked up uh and competed in the 10-meter air pistol category. And uh in terms of lessons, I think definitely teaches you resilience, discipline, and consistency, uh for sure. I think that's any sport teaches you that, uh, builds up that character within you. But what was what's a fun fact that I remember is that uh yeah, so uh I was not allowed to touch, and not only me, any new shooter, we we were not allowed to touch a pistol before one month. For one month, all we did was yeah, you know, the Indian the bricks in India, the uh the the and those are heavy, right? So yeah, you you you when you're a pistol shooter, you shoot with one hand, but generally your strong hand, which is a for the right. If you're shooting a rifle, then use two hands. So all to practice to make sure that the hands are stable, the the drill that we had to go through was that for an entire month, all we would do is just lift that brick in one hand and just stay there. And you stay there for like even 60 seconds is a lot, but you just doing that over and over and over again, and then then that leads your hand to be like rock solid. Because uh so then the logic was that the weight of the brick was equal to the weight of the pistol. So then when you train or do your shooting, your hand doesn't move at all because even like a millimeter or centimeter of move can lead to you hitting a bullseye versus a six or a seven. Uh, but that was a very interesting uh uh uh exercise that I recollect. And that also shows you, right? I mean, uh sometimes I think a lot of people think about lack of resources as a constraint. But I think uh most often than not, yes, it can be a constraint, but I think we are living in a world where you can transcend that. A lot of the times it's uh it's an internal mental problem that I cannot do that because of lack of resources. If you do the like things fall into place, you'll find the right people on the resources. So I think that's also one uh takeaway from that experience.
SPEAKER_03Interesting. Now I was smiling when you said that you wanted to join the Air Force, become uh become uh become uh FATAPAR, because I also had the same dream. I actually passed uh the uh the uh NDA uh exam, but my parents didn't let me go because I my eyes were horrible. I had to take minus two, minus 2.25 at that time, and I also uh and I also at that time used to stutter a lot. So my parents said like it's better that they would just stay at home, you know, it's uh it's funny. Yeah.
SPEAKER_02Well, I mean, uh uh Kharak Vaslam would have been fun. I mean, uh would you would you now as you look back, do you think how has your life turned out if you would have gone the end year out into the defense versus the life you have built now?
SPEAKER_03I don't know. See, like I like to live life in the fast lane, so become uh being a FADEP pilot would have changed my my thoughts, my mindset differently. Uh I didn't go down go down that route. I had a different uh experience. I would have never come to the US or started my own company, met my wife and had a had a family. So uh in the end it all happened for good, you know. Like either case, you know, like the way like you said, like the way uh life shapes you and it's what uh you make of it, you know, don't let life be the excuse, but uh be it a source of power for you so you can ship it the way you want.
SPEAKER_02Absolutely, absolutely. Since you're talking about power, huh?
SPEAKER_03Exactly. So uh outside of work, like uh uh what are you most curious, like curious about, like uh besides uh power gears, electricity, and so on? You know, like something that people would be surprised if they knew that hey, Bitun uh isn't to this.
SPEAKER_02I think I love water, uh anything to do with water, most of the times I relax by just uh going to swim. Uh so if uh I'm in the water swimming, it just I can relax no matter how much stressed out I have been. So that's a way for me to decompress. Lately I've been playing lots of pickleball. So if you're around and if you and we meet, we should definitely uh catch up for a game of pickleball. Uh and yeah, just I mean, just traveling and exploring. There's uh so many national uh parks in this beautiful country. So I am trying to take well hopefully all of them off. Um still, but yeah, just exploring uh uh some of the beautiful nature's national parks. There's a lot to do in Washington, so I just just go on walks and hikes and stuff.
SPEAKER_03Uh it is uh really beautiful, therefore show. You know, like in Dallas, at least we don't get that the mountains and the beaches. So we make up for it with some good barbecue.
SPEAKER_02Oh, absolutely. Nothing, nothing beats a Texas barbecue. I'm seeing that for sure.
SPEAKER_03Yeah, like you know, when I before I came to Dallas, I never had like I tried barbecue a few times in the northeast and I didn't like it at all because it was uh it was always sweet and there was always always a sauce to it. It's only when I came to Texas that I started to actually have barbecue.
SPEAKER_02Yeah, yeah. No, I I I moved from Houston, right? So like I can vouch I can vouch that the barbecue tastes different in Texas. It outside of Texas, it's not that good.
SPEAKER_03Not the same, yep.
SPEAKER_02My my personal opinion, I'm sure. Uh don't want to project it out there, but that's based on my limited experience. I think Texas is definitely the best place for barbecue.
SPEAKER_03Uh it's if uh it is too late now. Uh uh uh the cat is now out of the bag, and we all know that Texas uh barbecue is the best. So uh Vitun, like to tell us a bit more uh about your book. Like when is the book coming out?
SPEAKER_02In in in the next couple of weeks. The next couple of weeks, it's the final uh stages of draft. And uh uh for sure, like I said, it's uh it's to let the audience know about the uh the different constraints, the the like bring and tell them how and this would appeal to a lot of real estate investors because this is a lot of this is a lot about real estate investing, but it's real estate investing with a caveat where it you just can't just uh buy real estate and it all will be good. You have to buy a land, you have to buy a land with different constraints. Like I said, no floodplain, no chemical issues, uh too not too much grading, uh not like uh not near airport. There's like a 50 different checklists that I have put down in the book. So once you identify those uh good land, then the issue comes of power and all of those things. But it it does sound a bit hard, but if you could do that, you can literally flip that land for 10x a time, the cost. So it's all about how do you how can you it's an upgraded version of real estate investing where if you can package certain things that is required by the hyperscalers, the the people like companies like Amazon, Microsoft, they don't mind paying a 5x, 10x premium because you're saving them time. So the and then we talk about how you can uh and people with different uh uh can invest into this infrastructure space with different check sizes, whether you have a few thousand dollars to a few hundred million, based on that different strategies are also listed out. So uh it's more of a technical investment book towards the uh world of data centers and infrastructure.
SPEAKER_03Yeah, I was speaking uh uh with my wife this morning, you know, we are we are talking about how uh AI is going to uh replace a lot of jobs, you know, and at that stage, you know, like till it creates new jobs, there will be some amount of uh of uh time period in between, you know, where it will be important for us as a family to make sure that we uh invest in the right places to make sure that we are prepared for what is coming. So I think your book will be very uh timely and I would uh definitely look forward to it.
SPEAKER_02Thank you, Vishan. Absolutely. Looking forward to it. I'll I'll uh I'll uh make sure to let you know once it's in the final stages and uh yeah, we can uh happy to s do what I mean it's not clear, but if there's something I can uh or some kind of offer we can create for your viewers, happy to explore that as well.
SPEAKER_03Absolutely. Yep. Uh I was gonna uh going to ask that, but I'm sure like uh because the the book is not yet out, so I don't know like if you had anything in mind, but definitely.
SPEAKER_02For sure. Yeah.
SPEAKER_03Well, this was great, Pitun, and I really uh enjoyed our um our conversation, and I'm sure our listeners uh uh will also love it. Thank you again for being on the show.
SPEAKER_02Thank you so much, Vishad, for having me. Thank you. Have a good one.
SPEAKER_03Thank you, you too. This brings us to the end of another episode of Musings from the Cyber Trench. I'll see you on the next one.
SPEAKER_00Musings from the Cyber Trench monitor, proponents belong to the counter tank. It's like the active weak and collected with it.