← Oil and Gas Upstream
Oil and Gas Upstream
How BrightAI Is Transforming Critical Infrastructure with Alex Hawkinson, Founder and CEO | Ep 344
00:32:27
Loading summary
Transcript45 lines
- [00:04]
A
Oil and gas production is the union of natural systems with advanced science and complex engineering. Smart people across the globe create this remarkable place we call Upstream. And each day brings a new challenge. This is the Oil and Gas Upstream podcast where we look at how these systems come together and learn from the people who make it happen.
- [00:30]
B
Welcome to Oil and Gas Upstream. I'm Elena Melkert, your host. Some of you know me as the former Director for Oil and Gas Upstream research at the US Department of Energy. I retired from the doe, founded Energia Consulting and joined the Oil and Gas Global Network as a podcast host. I'd like to put a shout out out to my sponsor ifs. IFS supports the unique needs of the oil and gas industry. Learn more@ifs.com and now I'd like to introduce today's guest, Alex Haakonson, Founder and CEO of Bright AI. Alex, thanks for joining us today.
- [01:06]
A
Thanks for having me. I'm excited about the conversation.
- [01:08]
B
Oh yes, absolutely. AI is great, but unless you can make money through efficiencies and other things. Right. That's what the true benefit is for it. But before we go down that path, let me just refer a little bit to your bio. It says that you have over 25 years experience in IoT AI and SaaS and Cloud based technologies. That's quite a bit in that AI space. But tell us a little bit more about yourself.
- [01:36]
A
If only it was just 25 years at this point. But I need to update. That's how it goes.
- [01:42]
B
So what is your true number? What is your true number?
- [01:44]
A
I don't know. I graduated from college in 1994 so just as the Internet was emerging and I've been heads down as a entrepreneur and sort of scaling executive since then. So it's been a while. I can't believe it's been a while.
- [01:57]
B
Yeah, you're in the 30s now.
- [01:59]
A
It's pretty close. Yeah, exactly. I don't know. I've been a technologist my whole life. I went to Carnegie Mellon, came from a family of entrepreneurs and graduated just as the Internet was emerging. I, I went to cmu. I was fascinated with AI even then. A guy named Geoffrey Hinton, who's one of the sort of grandfathers of AI, if you want to put it that way, that sort of invented a lot of the learning methodologies that current systems are using, had a program at Carnegie Mellon back when I was there and fascinated from the beginning. But you didn't have the compute to do what you can do now. Long entrepreneurial career. I the Internet was just emerging web browsers and things when I graduated and I fell into a long entrepreneurial career. So I been a co founder in eight companies. I've run four of them as CEO and they've grown up with the Internet. So a couple of SaaS businesses, first a web development company, then a couple of SaaS businesses and then for the past 20 years I've been heavily turned into the physical world crossover space. So how do you bring intelligence and software in a messy physical environment? So before Bright AI, I founded a company called SmartThings. It's one of the bigger consumer IoT platforms in the world. It's several billion connected devices, 500 million homes. It's all over every country on earth that's got sort of modern civilization.
- [03:18]
B
So what does it do for a person at home, I mean is that they'll rent your house.
- [03:22]
A
It's the backdrop for a lot of smart home technologies. Samsung acquired the company down the line and they're about half of the devices. But it's a big developer ecosystem of connected devices in homes from lights to locks to thermostats to everything in between. Had more than a million developers create integrations and it's a backdrop. But the bad side of it is smart home technology is not for everybody. My wife is not a huge fan of things that I along that path. SmartThings is a well known and global journey. But turn my eye retired from SmartThings in 2018, turned my eye to okay, what are the most important problems in the world and it doesn't take long to arrive on. Energy is the number one building block for civilization and thank you for saying that.
- [04:03]
B
People don't realize, they think it comes from the wall.
- [04:06]
A
No, it's so founded Bright in 2019 and we're seven years into the journey now and it's of course these underlying technologies have all matured along this course and it's this intersection of physical AI is this intersection of a lot of the different pieces that have come together to make it feel like a new industrial era. But anyway, that's my background. Sit on a number of boards of directors, public and private and love, I love working with great industry executives to make transformations happen.
- [04:31]
B
Excellent, excellent. So you had an entrepreneurial spirit and you had a new and growing arena in which to express it. And it sounds like you were able to express it in many ways, at many times in different ways. And here we are, you're in energy, so that's wonderful. That whole journey gave you insights and experience that you're applying here. To the energy sector. And we love that. We love that. Tell us a little bit about your organization, Bright AI.
- [05:00]
A
So we are. Bright AI is again, we're in our seventh year now, so we deliver physical AI, we say for the world's essential services, but it's focused on all those things where civilization ends if they don't work right.
- [05:11]
B
And it's a. Oh my, that's a dramatic statement. What does that mean, the world's gonna end?
- [05:17]
A
Well, no, I mean it means again, if your smart lighting system at home doesn't work right, it's okay. But it means like all of every part of modern civilization really depends on energy and electricity and water and waste management and transportation of these things. So broad AI has a, I think the world's leading physical AI platform bringing, we'll get into, I'm sure what physically I is, but bringing that transformation of that into these industries that all civilization depends on. So we have a platform that we call stateful that lets you retrofit these existing heavy infrastructure environments and bring them into the age of physical AI. And it's an amazing time to be alive. It feels like the biggest, certainly the biggest change of my lifetime where you have an opportunity to completely transform productivity and default resilience and maintenance capex and quality work and safety and sustainability and all these things at the same time. So that's, that's what Bright's, Bright's focused on. So we can dig into the details as you want.
- [06:21]
B
Okay, so now it's making a little bit more sense to me. When you said physical AI, I thought, okay, we have a lot of pipe that interacts with rock and fluids and for oil and gas, that's what that is. And wouldn't it be great if we could control some of that with AI and the like? So am I getting warm? Tell us what that is.
- [06:41]
A
I wasn't from infrastructure before. I love learning. And coming into the company you could see that operators in these key sectors, they work in kind of a standard way in a sense. You have what we think of as observability. So they have infrastructure. You're trying to keep that up and running and make it work as well well as you can. And observability is a huge challenge. Like what's going on? Is there a leak in the pipeline now? Is the well performing well? Is there a safety issue at a given spot? And you can go on and on, but there's observability where you need to know what's going on and then there's decision making. Obviously you have limited information, so you have to make choices about go break if something's broken, you go fix it, et cetera. But hopefully it could be preventative and condition based on. And then there's action. You go out and take action in these environments. And the traditional way of working is very, there's very limited observability. So even in these heavy SCADA systemed places and in the most advanced oil and gas operators, they're very lean forward like still having to send people every day to these facilities to walk along. And why is that? That's because you can't, with the existing sensors and technology, you can't observe what you need to out of those environments you have to send is very labor intensive. People are fallible, they have to train them really well. It's tough environments, you don't necessarily take the best notes. So you get limited information, then that comes back in to make decisions around and it ends up being very reactive in a lot of cases, like there's unexpected outages or otherwise. And then the action itself is dirty and dangerous. Tough working environments. And so if you don't know what's going on in a place, when you send somebody, they have to be really highly trained to avoid real issues and otherwise and not be in danger. And so when you should think of it as physical AI, what it does is it transforms all those three things. So you can take observability and instead of having to send the person, you just know at all times what's going on. And you use sensors and robots and all sorts of cool technologies to do that. But if you solve observability, you also have much more information. And you can train modern AI models to predict issues before they break and before. And you can send the right guy at the right time knowing exactly what the right part is. And it's surgical action instead of this broad based dangerous action. So that's the spaces and we can get into the underlying technologies, but you can transform observability with things like that, sensors, robotics, those things, you get much more information and you can build models that can do better than any previous system at predicting the future. And then you can take surgical action just when you need to. And the outputs of that are all the things we talked about. We have tools for all those things. Yeah, it's a big platform.
- [09:24]
B
That's the business, right. It's one thing to drill the well and get the production and start making money and providing energy for the world, but then things get old or things just change. When you Have a lot of mechanical systems that are moving. There's literally a lot of moving parts and they all have to work in symphony in order for you to realize the reason results that you're expecting. And so, and some things you can see with your eye and some things you can't see with your eye. So when you say observable, you're not just limited to things that we could actually see, but you have other sensors and other ways to detect changes in the wrong direction from other things. Tell us about some of those things.
- [10:06]
A
Yeah, so it's interesting to dig into observability, decision making and the action. And there's changes in all places right now which are driving these Again we're on average seeing 2 to 300% like productivity gains. We're seeing huge maintenance capex changes, we're seeing production throughput increases. So it's important to get into the benefits on all of them as well. But it's imagine you're right, you have SCADA systems in a lot of oil, in oil and gas for things like what's the pressure on a pipe or what's the current being being taken out of a pump or other things like that. But it turns out those don't give you all the pattern that you care about. So let's say in a pipeline you're going to have a leak every month. In a facility you just don't know where. A leak starts as a pinhole leak and then develops into something catastrophic. But the pinhole, you can't even sense it on the pressure sensors because the pressures are so high generally. So imagine with next gen sensors you can do things like add vision at the edge or audio or vibration analysis or these things that took like a technician being there to walk along and do it. You can now do with these next gen technologies. Like I don't want to pitch things specifically but to take an example, like a sensor that you can just stick into a place will last a decade on its own on the battery. And that can give you those modalities that you had to send a person before with a bunch of instrumentation to go see. And so you can imagine that being like detecting these things that were previously unobservable. So sometimes you do that with leave behind technology. Sometimes you really need the vantage point of say in the air or underneath something. And you use robotics to bring those multimodal sensors into place. And in the worst case scenario you send the people. But if you send the people, they're not just notepads and things. You can Equip the people with wearable edge sensors and AI where you can collect the data at a much different level. So in observability, you're right. There's new types of sensors that didn't really exist before. And then there's the way you process those at the edge, where you can gain these patterns to do what the edge advanced technician could do before, but you can do it all the time now. What does the pump sound like? How's it vibrating? These things that a human operator would have been required to go see before, now you can do remotely, which is amazing. In the observability front, one of the
- [12:24]
B
things that we're noticing, for example in the subsurface, is that since we did not have ways to analyze particular types of data, we just had to ignore it. But now that we can analyze, now we're interested in this, I don't want to call it non traditional data. It was always there. But are you finding that because you can now you've got new ways and more sophisticated or advanced and give you that part of the information you need? Tell us a little bit about that.
- [12:51]
A
Yeah, so there's cases, I think, of these observability problems of there's different reasons you have the gap, right? Sometimes you don't have the sensor type. It just hasn't existed yet. And it's becoming prevalent now, say a thermal camera or a molecular detector or some new advanced sensor type. Then there's something that's perception. So sometimes you had the sensor, but it was so much data that you can't even record it or transmit it to the cloud. That's we've seen very frequently. And there you have new, we have a toolkit for basically processing that data at the edge on the piece of equipment where say vibration data is like that. If you sent it all to the cloud, it's an enormous amount of information and it's impractical to do that. And so they just, they might have the sensor, but they have to just basically ignore it. It's only for the. When the technician goes, now you can take. And we have an edge AI hub that you can leave behind in a spot and it can see the patterns in that data right there on the piece of equipment. And you don't have to send everything up to the cloud, you just send it the pattern that it sees instead of that huge volume of information. So you have new technologies that are unlocking sensing and perception. Then there's AI models for decision making. Okay, let's say you see the basic pattern. How do you predict the future in some way? And I think we've all witnessed how much AI has changed over the last few years, where if you have observability, you can train a model. Now with architectures we didn't even have a few years ago to really see those subtle patterns that human operators could not see before. And so but you have to solve each of these things in a layer. It's like sensing, perception, prediction and then closed loop action is another thing is like when you send the person out, did they solve the problem or not? We again, we solve all of those types of gaps.
- [14:39]
B
So this notion of progression from one capability or insight to the next, both in terms of sequence and in terms of sophistication, deeper understandings, deeper insights, is that something that. I'm just trying to connect that with the fact that you watch the Internet grow and become what it is over time. Is there some relationship between those two things that made you think in that way that allowed you to move?
- [15:05]
A
I don't know if this gets to your question, but as you get older, you want to have an impact in the world and you look for the biggest problems you can go solve and contribute around. And to do what we can do with physical AI now, like it wasn't possible even just a few years ago. And it requires multiple layers of new technologies, new sensing types like what we were saying, new edge computation that didn't exist before, new architectures for processing that information to find patterns with AI and making that all economically feasible. So it's not going to bankrupt a company to go put it in place. But there's these moments in time as an entrepreneur, you get to recognize when thumb thing, that was a key constraint before, is now no longer a science project. Now it's applicable. And we're right at the juncture where for physical AI you need all of those things. You need edge computation sensors, very robust IoT kind of equivalents, you need these pattern recognition sort of capabilities. You need all of that stuff together. So I would say two things. As an entrepreneur, you learn when things are doable. And that really wasn't what we're talking about even a couple of years ago. And so this is the moment nobody should feel dumb about as operators about not having a fully deployed these things. It's just now is the moment for it. But it's also you hone in, you hone a sense of like how big a problem space is. And again, it's, you look at energy and you're talking about 10 to 20% at least of the worldwide economy in terms of the underpinnings of civilization. And so it's hard not to be excited as an entrepreneur about oh, you can go impact this thing that all civilization depends on too. So it seems more consequential than the Internet as well.
- [16:48]
B
Yeah, yeah, I wonder. So the data analysis, the observable parts and then the processing parts, that is all dealing with infrastructure that exists and the way we do things. I want to flip the question have you think about from the end, will it also give us insights as to how to do it better, more efficiently with fewer moving parts? Tell us that story. I think that's. And where I'm coming from is the notion of energy poverty. Percent of the world does not live the way we do because they don't have what we have in terms of energy. As you shared correctly, it's fundamental to your way of life and your being able to express all your personal talents. If you don't have light at night, you can't even become a better reader. Or it's just these things that are holding people back that technology can break through for us. So if you could bring power to people in these remote areas more simply and more directly because of insights. Because we don't have to do it the way we do it now. Tell us some of that. Tell us that.
- [17:48]
A
Yeah, I think there's multiple facets to that. And the answer is yes, at a high level it definitely gives you the insights because if you weren't observing in the past, you, if you send the guy once a day or once a week to go observe it. And there that one time, everything that happened in between you miss and you don't know what was occurring and you could infer it after the fact. So when you don't have observability, that's sophisticated. You are missing a lot of insights that exist in that in between time, so to speak. When you add that in and then you pile that into a modern AI engine that can predict issues, a number of different things happen. So a you avoid downtime in the first place because you can see we've had customers. Take Kodiak Gas Service is a good example. Like a great company, a leader in kind of gas compression where we've seen that you can go from problems that were previously reactive, like you only find them once something's broken to seeing those issues like from hours to months in advance through prediction. And when you can. Everybody knows that when something catastrophically breaks, it's much more expensive than you. You get in front of it, and so on. It's on the basics of having observability, predicting the problem, getting in front of it. You avoid downtime. There's another thread in this, is that you, it turns out that a lot of infrastructure in the world is running at, let's say, because of lack of observability. The vendors will say, oh, you should run it at 80%. Let's say when you have foundational AI models that know your infrastructure is healthy, you can actually push them to higher limits. You can push to 90% that figuratively. And so we've seen example customers where you can increase production by 10 or 20% based on having these physical AI things where you know it's healthy, you can really push and you can avoid catastrophic downtime on the other side. So you have these like wonderful effects for supplying energy, like already just built into the existing way of doing things. Like if you add observability and decision making. Now to get to your question on top of that, when you see suddenly you have observability, you can see, oh, it's this model of compressor that breaks down faster than this other model and here's how it fails, or it's this part type of pump, or it's this type of pipeline segment in these weather conditions that breaks down more quickly. That then gives you insights to say, oh, let's redesign, let's redesign for simplicity, let's redesign for reliability. Because you, you see what all the common issues are at a much, much deeper level than you ever did before. And we're seeing the front end of that design curve, which is really exciting too, as you can design then for next end reliability, design for being able to run things more aggressively, et cetera too. So it's really exciting when you're talking about hundreds of percentage points gains in productivity, massive changes, eliminating, let's say, half the unexpected downtime or more. And then not only that, like increasing the amount of production flow you can get from things because you know it's operating healthily and you have the confidence to do that. Like that alone is existentially powerful for these companies and offices.
- [20:58]
B
Absolutely, absolutely. And does that also extend to design that you could design something more efficiently? Because now we know things we didn't know before, we don't have to do it the way we did it 10 years ago or 50 years ago. There's some things that haven't changed design in 100 years. And especially in the oil and gas business.
- [21:17]
A
Yeah, absolutely. From the data, everything flows from observability. When you fix observability, great operators, they get all sorts of new insights and it can help you in all those regards. It'll also help you with. Okay, we should. How do we eliminate with generative design like these key top 10 problems that we have. Right. And get into that and the electrical power industry. Take gag with wind turbine like with turbines and wind. These turbines for power generation in general. They were an early company that got in front of things enough, you could actually design them for reliability on the other side and participate in that. We're now seeing that wave across all of these other sectors because it's much more practical. You don't have to have billions of dollars in machines now. You can retrofit pretty inexpensively existing infrastructure to get all those insights falling out of it. So really excited. Yeah, that's a lot of money. There's a lot of money involved. It's amazing. I grew up, I used the analogy of the whack a mole game at a carnival where you hit the. Everybody knows this from kids. You'll hit the one thing and something pops up in another spot and you're trying to whack a mole down. The beauty of physical AI is it's not whack a mole. You both increase productivity, but you also increase quality, you also increase uptime, you also increase sustainability, you also dramatically improve safety and all these things as well. Because you're not sending human crews aligned into a lot of situations. You're sending them out surgically. They know exactly what's going on, you can guide them, et cetera. So it's like literally everything improves all at the same time. Which is why it feels like such a. I think that's just to put words to why people, I think are feeling like AI is the biggest deal of our lifetimes in terms of the level of change. Literally everything changes by these massive material level. So you make more money. People are also way safer. And it's by the way also more sustainable and also use less maintenance capex and parts. It's great. It's like literally it hits all those
- [23:14]
B
factors and then an arena that's I don't want to say especially vulnerable, but has a lot of extra moving parts is offshore like the Gulf of America there, you know, with all of the water and as we move deeper and deeper water and higher pressures from bigger prize. And then there's the aging infrastructure that's out there. That's. That's what was motivating my question. Because there's just. When we talk about conventional Production as opposed to shale development. The recovery factor is, it's higher than shale, but it's still not 80% even that's 30, 40, 50%. 50% is a big number. I don't think we get that far. I want to go back to the future and think of ways that we could analyze previously gathered data that we just ignored because we didn't know how or we didn't think we could care about it or whatever. Because the infrastructure is such a huge cost, but you can't get the production without it and the price of oil is not tied to what it cost you to develop it. Talk about that. Do you have some trends maybe or some thoughts about the offshore potential?
- [24:19]
A
I think you're going to have more expertise in this specific regard than I do. We have again top 10 sort of customers worldwide, producers in all of these different spaces. But it's another example where you have an enormous number of observability problems, right, that come from some cases like you don't have the sensor technology, there's new sensor technologies emerging. Some cases that perception issue where it's so much data but you it's impractical in this remote facility to send that all up to the cloud to hopefully sit somewhere else. So you have the things like the edge AI capabilities to process the data in situ. You have these in this enormous amount of data. It is absolutely remarkable. Everybody's felt this AI capability on this path that it's on where fed the right information, it can predict things better than any of these previous human systems. Like where you have a bunch of guys typing out the rules effectively for heuristic systems to manage a plant or manage a facility or one of these things, AI can see patterns in data that is just beyond human sort of comprehension. And so applying that to the existing data streams and having closed loop observability where you can actually fix things. So put all that together and something we haven't really talked about yet too is how robotics are changing. That's just another facet of using edge AI on a thing with motors to bring the sensor where it needs to be and maybe even take an action. And in these dangerous environments like an offshore rig and the rest of it. Add that to the mix too, of eliminating a lot of the places you'd have to put a human in the loop with next gen autonomous edge AI and robotics and you have the solution to a lot of the biggest problems that plague it. I would say if you want broad brushstrokes like in all environments, not just one subsector in all environments. We've seen on average 2 to 300% productivity gains. Think about that. So the same crew is like if you have an aging workforce and it's very difficult and to protect and recruit these people, that same crew can do three to four times as much work. You have maintenance capex drops of 25 to 50%. If you're spending $500 million a year in maintenance capex like several hundred million dollars a year in freed up capital that you can apply to new systems and engineering. And the other things you're talking about production throughput you have gains of again like not just out of all the avoidance of outages, but then you also have the production throughput on a healthy system increases by 10 to 20%. And then on top of all that things run more efficiently so you avoid sustainability issues and leakages. And then on top of all that you look at these critical industries and you say I probably have an injury in all of our sectors, not just oil and gas, but you probably have a frontline worker get injured every second. Most of those injuries are preventable and you avoid those too. So it's like it hits all the factors. So I think those same factors apply to offshore things where you can like okay, people are scarce to send out there. It's very difficult labor. What if you could get three to four times the amount of productivity out with lower training requirements. You have maintenance capex huge. What if you can do that better? Better if you have higher throughput production, if you have higher safety, you solve all those things at once and suddenly you can produce a lot more energy because it's much more capable of. You can take all those saved benefits and apply them to additional infrastructure and ramping worldwide.
- [27:46]
B
Absolutely makes you want to go out and design a new sensor so I can find something else. I don't know.
- [27:51]
A
And it's amazing happening. The beautiful thing about what we're doing is we have a platform. That's what some of our customers love is it's end to end. We have one of the top 10 oil and gas producers sample that will announce soon. But it's love like they're not just oil and gas. Like they have their own power substations, they do their own underground utility mapping, they do their own all sorts of different aspects. And the fact that you can fold in all of their operations into a single operating platform with physical AI with, with what we're doing is really exciting. And then we can benefit when there's new sensor types we can just ingest them into the platform. When there's new robotic chassis for doing new robotic work, lots of innovation happening there. We can just tie that new robotic chassis into the platform. So it's, it's a fun time to be alive.
- [28:34]
B
It is a fun time.
- [28:35]
A
So much innovation.
- [28:37]
B
We are almost out of time. Was there anything I didn't get to ask you or that you wanted to be sure that we share especially about Bright AI?
- [28:45]
A
I guess there's probably two things. Of course with Bright we're at the place now our first customers when we got started took two to three years to implement to get to the benefit levels we're talking about. Now we see that same type of dynamic within 6 to 12 months. I just encourage operators to now that we've got the picks and shovels of physical AI, meaning these next gen sensors and these wearables that you can put on the frontline workers and the robots and the other things, it's a moment in time where if you want the competitive advantage of pioneering, now is the moment and just encourage people to engage. And we just love going out in the field and putting the hard hats on and rolling up the sleeves and showing how that impact can happen very quickly. But that's of course the thing there and we're. I think we're the best funded in the space doing really innovative things here. So we'd love to talk to all of your audience members but I guess the one insight we didn't talk about separating that from even Bright is also think about the edges. It's funny observability problems like we're solving, we've seen this really interesting phenomenon where they're not just in one company. There might be the intersection point so let's say where the gas supply meets the manifold for distribution and one company owns one side of the manifold and another company owns the other side. And there's places where there's a problem there of you can solve it by knowing hey, you have this gas to send down the pipeline. This recipient maybe has health issues of their equipment where they can't receive it in one path and they can in another. I'd encourage people to even think about it's not just your company too. It's actually can affect your supply chain and like where you're intersecting in the industry and we've seen there even another level of the kind of squeeze and value and profitability and other things that can come out of things. So it's just lots of exciting. Everywhere you look there's these opportunities and
- [30:35]
B
that whole interface, intersection and seamless connection is a way to put it. We see that clearly in human systems. You put one tribe together with another tribe. So of course anything that we design, design and want to interface with something else is going to have that same challenge. Maybe if you come up with solutions in mechanical systems, we can apply it to human systems and we could get along a little bit better, have easier, better ways to communicate. Oh, this is fascinating. I'll be sure to put connections links to your website in our in the show notes for this episode. Oh my gosh, it's just been so fascinating talking with you. I especially love the part about how you've watched the change of just our computing capabilities and understandings and then the interface with the human thinking as well over time. That's just exciting.
- [31:24]
A
One more thing not to cover.
- [31:25]
B
Oh yeah, no please.
- [31:27]
A
Yeah, I'll say for your sponsor too. We've actually done a lot of great work with iFest, so it's a great company and lots of intersection points for those of your customers that are using them to leverage physical AI with companies like us too. But anyway, thank you for your time.
- [31:41]
B
It's a pleasure to Alex Hawkinson, Founder and CEO of brightai. Thank you so much for joining us today.
- [31:48]
A
All right, thank you Elena. Look forward to talking again in the future.
- [31:51]
B
And thank you everyone for listening. This is Elena Melkert, your host for Oil and Gas Upstream. More next time.
- [31:58]
A
Thanks for listening to oggn, the world's largest and most listened to podcast network for the oil and energy industry. If you like this show, leave us a review and then go to og oggn.com to learn about all our other shows. And don't forget to sign up for our weekly newsletter. This show has been a production of the Oil and Gas Global.