Transcripts
What You Missed On The Vector Episode 24: Powering the Future: AI’s Role in Space Innovations and Operations
Written by: Morsiell Dormu
In this engaging episode of The Vector, Kelli Kedis Ogborn, Vice President of Space Commerce & Entrepreneurship at the Space Foundation, hosts Lorenzo Feruglio, CEO and founder of AIKO, to explore the transformative role of artificial intelligence (AI) in space operations and innovation. With over eight years of experience in AI-driven flight software for small satellites, Lorenzo shares insights into how automation is revolutionizing space mission efficiency, reducing risks, and unlocking new economic opportunities.
Highlights include Lorenzo’s insights on:
AI-Driven Satellite Autonomy & Mission Optimization:
Lorenzo discusses how AIKO enables satellites to make autonomous decisions in orbit, such as identifying unusable images due to cloud cover before transmitting data, significantly reducing inefficiencies in satellite imaging operations.
The Evolution of Space Operations & Predictive Maintenance:
AI-powered predictive maintenance is shifting space operations from reactive to proactive, allowing operators to anticipate failures before they occur—improving system reliability, minimizing downtime, and saving costs for satellite operators.
The Future of AI in Space: In-Orbit Data Centers & AI-Integrated Networks:
Lorenzo envisions a future where entire data centers operate in orbit, allowing AI to process satellite data in real time, reducing latency, and increasing operational efficiency. With expanding inter-satellite connectivity, AI will enhance automation, communication, and decision-making processes across space missions.
AI’s Role in Space Cybersecurity:
With increasing threats to satellite networks, AI-powered anomaly detection can identify cyber threats before they impact operations. While AIKO doesn’t directly provide cybersecurity services, their AI models contribute to early detection and mitigation of cyber risks in space systems.
Challenges in AI Adoption for Space Missions:
Despite AI’s potential, industry hesitancy and legacy infrastructure slow adoption. Lorenzo emphasizes the need to educate companies on AI’s business benefits, such as cutting mission analysis time from 10 days to a few hours, making space operations more efficient and cost-effective.
The Long-Term Vision: A Fully Autonomous Space Economy:
Lorenzo sees AI driving greater accessibility to satellite services, where businesses and individuals can easily task satellites on demand. Automation will streamline interactions between Earth-based users and space-based assets, unlocking new commercial applications.
This episode provides a fascinating glimpse into AI’s expanding role in space operations, from optimizing satellite efficiency to shaping the future of autonomous space missions. Whether you’re an industry leader, space entrepreneur, or AI enthusiast, Lorenzo’s insights offer a roadmap for leveraging AI’s full potential in the space economy.
Stay connected with the Vector Community! Visit us here to sign up for the Vector Newsletter for quarterly updates, news, and stories, or opt into monthly updates about The Vector Podcast, featuring thought leaders in space, entrepreneurship, and innovation. You can even suggest topics or speakers for the podcast. Don’t miss out—subscribe today!
Episode Transcript:
Kelli Kedis Ogborn:
Hello everyone and welcome to the Vector where we discuss topics, trends, and insights shaping the global space ecosystem. I am your host, Kelli Kedis Ogborn, and today’s discussion is centered on the role of artificial intelligence in space operations and innovation. Varying forms of AI are expanding at an unprecedented pace across all industries and its utilization to gain efficiencies, reduce risk, and enable new discoveries, make it a really powerful tool within the space ecosystem. From autonomous spacecraft, AI driven mission planning, and real time data analysis and robotic exploration, we are seeing AI act as both a guide and a problem solver. My guest today is no stranger to the topic as his career includes over eight years of experience developing flight software and AI applications for small satellites. Lorenzo Feruglio is the founder and CEO of AIKO, which is a company that was established in 2017 and has become a leader in upstream space applications with offices in Italy and France.
Under his leadership, Aiko has grown to over 40 employees and continues to innovate in the field of autonomous space systems. Throughout his career, he has contributed as a visiting researcher at NASA’s Jet Propulsion Laboratory and at the Massachusetts Institute of Technology MIT, working on small satellite mission design and telecommunication architecture modeling, as well as gain professional experience at SES in Luxembourg, which is a global leader in space telecommunications. He holds a PhD in aerospace engineering, and his research has focused on artificial intelligence to enhance the mission autonomy of small satellites. Lorenzo, welcome to the show.
Lorenzo Feruglio:
Thank you very much. Thanks for the opportunity.
Kelli Kedis Ogborn:
No, we’re really, really happy to have you and especially because you’ve been engaging in this topic really from the ground floor. Eight years doesn’t seem like a long time, but in AI world it is a very long time. So before we dive into really the bulk of the conversation, can you tell us a little bit about your company and your work and how you were bringing artificial intelligence into your operations?
Lorenzo Feruglio:
Yeah, absolutely, absolutely. So I think the most simple way to explain the company is that we are really an automation company, and as any automation company, the goal of such a company is to drive down the cost of specific industrial processes or to increase the capabilities of a given robotic system. So this is a really general description that I do at AIKO, but it’s really easy to understand the nature of the company. Now, the specificity of our company is that we work in a very crucial domain and vertical, which is the space economy, and therefore the type of automation and the type of value that we drive is very peculiar and very, very complex for this domain. And in fact, driving and delivery automation in the space economy really means that we’re delivering complex automation and therefore we use a lot of AI in our business and therefore we are positioning ourselves as a leader in the AI for the automation of operations.
Kelli Kedis Ogborn:
That’s really helpful and I want to break it down a little bit further. Maybe not so much about what your company does, but more just the use case of artificial intelligence within current space exploration and operations. Because AI is in a lot of ways a blanket term for a lot of different flavors within that of how people are actually using the tool and using the underlying technology. Can you just give some examples of the way that the industry is currently using it?
Lorenzo Feruglio:
Absolutely, absolutely. So maybe before we jump into specific example, just to give you an understanding of the type of automation we deliver and where do we actually deliver it, because satellite admissions are kind of a complex, let’s say scenario in this way. So first of all, the company AIKO, as we are really delivering complex automation, we work in two directions. The first one is software that flies directly on the satellite in orbit, so it’s embedded within the satellite that flies in orbit. And the second direction is in the mission control center, which is the big room where typically you see in the movie the operators, the classical Houston, we have a problem scenario. That’s where it happens, and those are the two main points of a space mission where we work. Now that we understood this one, just to give a couple of examples, onboard the satellite, one of the leading technologies that we have is actually giving the capability to a nerd observation satellite to understand the content of the pictures that the satellite is taking to make actionable decisions on these images.
And to clarify, I always give the example of the cloud detection. Today, satellite imaging satellites deliver nine out of 10 images that are not monetizable on the ground. So this is really a striking number to me because it’s super inefficient the chain in which the satellite is able to capture images and then deliver these images. If nine out of 10 are not monetizable, one of the biggest reason is that actually they are covered by clouds. And what we give onboard the satellite is the capability on the satellite to understand that a current image would be completely useless because of the cloud coverage. That would be pretty high and nobody would actually purchase that image on the ground. And we give thanks to our deploy models, the capability to the satellite to say, no, this is not a useful image. Let me schedule the acquisition for another time.
And this is a clear example in which we deliver value onboard the satellite. If you look at the ground in the mission center, one of the key topics we work on is predictive maintenance. Today we changed the approach of the operators that typically react to a failure that has been detected. So typically they scramble to fix the failure in a quick way. In our approach with predictive maintenance, we anticipate the detection of the prediction actually of the failure, making sure they can prepare even to react to the failure when it will happen or even to avoid the incurrence of the failure.
Kelli Kedis Ogborn:
That’s really interesting. So going back to your first example of being able to give better fidelity to the type of image that the satellite is capturing due to cloud coverage, is this then from the decision-making point on the ground, reducing the volume of images that it sends down or how does that work for the human aspect to actually make their job easier? And to your point, the monetization, because that’s the goal of a lot of these smaller satellites,
Lorenzo Feruglio:
Correct? Correct. And I would say that the main KPI is really to increase the number of monetizable images that these companies actually receive on the ground. So it’s not really about reducing the overall data which is received, but it is actually making sure that this data that is sent down is actually monetizable. And if you think about that number I gave earlier, nine out of 10 not monetizable, if even we reach a point where three out of 10 are not monetizable, this is a strong improvement on their business model, on their value chain. Some companies do actually pay money by renting the equipment, the antennas and all the infrastructure they pay depending on the amount of data they download from the satellite. So there could be a monetary value even in that direction. Basically we reduce the overall amount of data, but to us driving the value up of the space economy is really making sure that the data is actually meaningful, that we get as much data as possible is actually meaningful.
Kelli Kedis Ogborn:
How does your technology, I’m trying to think of how to phrase this. So how would it expand or support or need to adapt given where the space economy is going, especially with the proliferation of more satellites being put in low earth orbit and geostationary? I’m just thinking about future mission sets because you’re absolutely right that if we are going to be able to enable growth economically, the commercial viability of a lot of these images is going to increase, but there’s now going to be a lot more satellites. So what does that future look like for you and particularly around artificial intelligence systems to keep growing the monetization of space?
Lorenzo Feruglio:
Absolutely. I think that the future is really interesting because what we are seeing is a strong evolution of the whole infrastructure in space. So we see now concepts of full data centers being sent in orbit, the capability of hosting networks, of satellites that can share the data among themselves, and basically having a satellite that capture an image, deliver the image to a data center satellite in orbit and have the data set itself basically run the algorithms and run the processing. And this is, for example, a strong evolution because today we are limited to putting AI on a satellite that is supposed to take images. But if you think about having a full data center in orbit, this is a game changer not, and that’s one point. Another point of interest for us is the proliferation of the connectivity of the satellites. A satellite is limited by how many times and how frequently it can connect to ground.
It can send data to ground, but with the proliferation of starlink and other constellations where the connection between satellites is increased, we get to a point where the need of putting autonomy and AI on board the satellite is shifting with respect to ground. And this really depends on a few factors. The need of autonomy will always be there because it optimize the value of what the space mission is doing. It’s the allocation between ground and flight, which is changing. And ICO as a company is constantly monitoring this evolution to make sure that we correctly place our products in our capabilities. It is,
Kelli Kedis Ogborn:
Yeah. It’s interesting we, so the Space Foundation looks at economic data for the holistic year before and gives a quantification and ground station sort of year after year and one of the largest growing sectors primarily because of the necessity and the proliferation of satellites. I’m curious from your perspective, where does cybersecurity play into this? I agree with you that we are moving into this eventuality where we are going to have data centers in space and the need for broader connectivity and artificial intelligence to enable those connections autonomously or critical. What about the cyber component? Is that a concern and can artificial intelligence also help that run more smoothly?
Lorenzo Feruglio:
Well, cybersecurity in the framework of resilience of a space mission, the framework of avoiding cyber attacks and basically not limiting the satellite operational capability, it’s a really, really strong topic. The industry is investing a lot in this topic and making sure that the satellite technologies are keeping pace with basically the technologies that we see anywhere else in the world and making sure that they are protected by potential cyber threats, cyber attacks and so on, which are becoming a reality in these years in any domain, artificial intelligence can improve and help in the cybersecurity, especially threat detection. This is a very common technology in the FinTech domain in many other domains. And our technology, since we work on predictive maintenance and anomaly detection, can be applied in detecting anomalies that are generated by a cyber attack. So this is one of the direction we are looking forward to expand in the future, not really delivering cybersecurity services ourselves. We’ll remain an AI and automation company and making sure that our anomaly detection AI can actually be put in use to detect cyber threats and support on the broader cybersecurity topic.
Kelli Kedis Ogborn:
Can we expand a bit? When you talk about predictive maintenance and anomaly detection of space systems, I don’t actually know if I fully understand really what that means and how AI is going to grow that in the future. So can you talk a bit about Yeah,
Lorenzo Feruglio:
Absolutely. Absolutely. So to me, the first thing to keep in mind when discussing about detecting failures on a satellite, which is really the topic here, we’re trying to reduce the number of failures on a satellite because if a satellite has a failure, it’s likely to go in a downtime and if it’s in a downtime, it’s not operational and not delivering the service that it was supposed to deliver. So this is the overall goal, but you need to keep in mind that the satellite, it’s a super complex machine, it’s worth even hundreds of millions of dollars of euros, and it has even 10 thousands up to 50,000 sensors that are monitoring little piece of equipment, little components on the satellite. And this is a huge amount of information. And here the complexity is that today the operator is kind of left alone on the ground in trying to understand all the complexity and all this data and trying to understand where something is not going properly and how they can react
Kelli Kedis Ogborn:
Different systems and sensors.
Lorenzo Feruglio:
Correct? Correct. No, so the use of AI here is really to compliment and actually to support the operator in analyzing all this data and making sure that the operator can focus on the resolution and not really wasting time in skimming through all this data. And a classical example that I give here is this. Now the space economy today is really used at generating alarms. When a certain parameter, a certain piece of information is crossing a threshold, it means that they are somewhat used to be alarmed. If a parameter is going out of bounds, let’s put it in a simple way, but a lot of the anomalies, a lot of the failures actually happen way before and signals can be detected even before the parameter crosses the bound. And that’s what we do. That’s where AI comes into play because you can detect the patterns and signals that anticipate the occurrence of a failure before this actually triggers some sort of alarms, and that’s where we come in as added value.
Kelli Kedis Ogborn:
And how does the human in the loop aspect play into that? So is it more that there’s information delivered to the person and they make the final decision, or do you see an eventuality where the AI can also then help trigger something to correct and fix the problem?
Lorenzo Feruglio:
A hundred percent, a hundred percent. The vision of the company is not really to, well, the vision of the company is to completely automate the satellite operation per se. So it’s not just the detection of a problem, but it’s also the resolution of the problem. So you’re pointing it out correctly. The future for us is a set of AI capabilities to detect or predict some issues from contingencies on a satellite, but also to automate the response and the fixing of the problem itself. For us, the ideal scenario of a future is that our customers, they launch a satellite and they forget about doing operation at all. So that’s the goal that we’re
Kelli Kedis Ogborn:
Chasing. That’s interesting. I feel like you can’t have a conversation about artificial intelligence without having some sort of discussion about the trust factor or the comfortability of a system running and trusting the process. And I know that for what you describing and its operations and what it enables for satellites is slightly different than the eventuality that people think of Terminator in these other systems. But when you are working with customer sets, do you find that there is an increased trust factor in the ability to do the AI to what it says it’s going to do? I go back to your point about talking about how expensive these satellite systems are, right? And they are extremely critical to mission sets and obviously to the customer’s bottom line. And so do you see that trending in the right way where it will get to a point where people will just totally trust the system to be able to do what it says it’s going to do and think about other things?
Lorenzo Feruglio:
I do see this trend, but because in AIKO, but maybe other companies outside the space economy are doing the same, it’s really a matter on how you deliver in production, a specific AI model. The real complexity is not really developing the prototype. Now thanks to the courses and everything that’s available online, everybody can build its own AI model, but then when you have to put it in production to manage a mission which is worth a hundred million dollars, that’s the real challenge and this sort of operational framework to bring the model to evolve up to production, that’s where the trustability aspect coming. And for us, we work with our customer really in a synergy because they get to try the model first. They get to try the model without incurring in any potential risk for their mission. We monitor the confidence in which the model is able to deliver the results that the customer is expecting, and then we get to a point where the customer is ready to adopt it in production, and this is really important to drive and then have the customer follow you in this path of adoption in production of ai.
It’s really key.
Kelli Kedis Ogborn:
Yeah, I imagine it’s really important to have them involved and integrated now so they can grow with the models, grow with the process. It really is that linchpin that’s going to be very critical to enabling growth in a way and allowing humans and just people to focus on other things that are just as mission critical, but knowing that satellites are actually going to, or the mission rather around the satellite is going to perform as it’s going to.
Lorenzo Feruglio:
I can give you an example on this, which is really, really striking. One of our customer went from 10 days of data analysis time for a postlight for analyzing all the data received after the satellite was launched two, three hours. So going from 10 days to two, three hours is really a game changing scenario. You get so much more time to do other stuff, which is really a value for these companies. Yeah,
Kelli Kedis Ogborn:
And it helps close business models and probably new use cases that they never really had the time to consider before. When you think about future growth, so since we’re sort of going on this narrative path of space economy growth and commercialization, not so much how it’s being used now or how you use it, but when you look toward the future, how do you see AI influencing the commercialization of space and really the growth of the space economy? So you can answer this in a lot of different ways if you think it’s just a continuing trend or if there’s new utilization and use cases that may be coming in five years that you’re very excited about.
Lorenzo Feruglio:
Yeah, absolutely. So what I can tell you is what we are chasing as a company, and it’s really not even a matter of AI, but overall of automation per se. Because at the end, AI for us is the technology that we deliver to enable automation, but it’s automation. The goal for us and what we’re really chasing is really expanding the way common citizens or companies on the ground can leverage satellite services. Today, the space economy is really affected by an issue, which is it’s very difficult for businesses on the earth to interact with a satellite service. We use the navigation on our phones. We do a few other things, but it’s really a disconnected reality to me, and the role of automation, the role of automating all the aspects of the value chain can really get to a point where a citizen can really maybe from the phone task a satellite to get the image of their own house completely in an automatic way, whether today you need to figure out who’s selling the images, how can I request a specific image? They need to command the satellite a highly manual and complex chain, I would say. So the future for us looks like this. Now everybody on the ground, on the earth interacting with space without worrying about who do I need to talk to get something from out of the satellite service?
Kelli Kedis Ogborn:
I have so many follow questions around that because I agree. I think that that’s where things are driving, and you’ve definitely seen that sort of trend in other industries. One of the things that I talk about often is when people think about the evolution of the space, space markets and the space economy, it can draw a lot of parallels to aviation in the cell phone industry because it started really heavy government intervention, trended toward commercialization and then opened up and enabled all these new markets. And when you particularly think about the cell phone industry, you rarely use your phone for calls anymore, and it’s like all of the apps and the widgets and the other things around it, and what you’re describing is sort of that eventuality because then it is, it’s apps and connectivity and it’s more connected to the person. My question is the complexity of that because when you look at the security aspect, and I’m going to draw a parallel to the drone industry, so the technology around autonomous drones came to fruition particularly for military applications. Then once it became commercial, there was the realization that there were autonomous vehicles with cameras flying over private citizens’ houses, and then there was the regulation aspect within it. What else is involved in that aspect that could prevent that commercialization reality? Because I imagine a citizen calling up an image around a sensitive site or a military base could be a non-starter maybe.
Lorenzo Feruglio:
Yeah, so I think this is a very complex question. It involves a lot of different direction in which the space economy itself can evolve, but if I answer from the point of view of the common current understanding on the evolution, if you think about the air observation companies, so companies that are launching satellites for taking pictures and then selling the pictures or selling the insights that they get out of the pictures, the common understanding this year is that these companies need to evolve to become application companies. It’s not really just selling the data that you need to do to make sure that you close the loop on the space economy and you deliver value to everybody on the world, but you need to do deliver application. So they need to close the loop on serving the data the inside that is needed at a given moment through a set of infrastructure on the ground that is also up to the application on your phone, as you were saying. And I would expect that if this is the evolution that they will be in the earth observation market, then they will be regulated in a way where they cannot offer similar, they cannot answer all the requests they get, especially if they target sensitive areas, for example, and you have seen this in the drone market, in the commercial drone market. When you buy a drone, you have restricted areas embedded into the drone flight software. So this is an example.
Kelli Kedis Ogborn:
Yeah, I like that answer because it is driven out of necessity from a business model, but also just the evolution of where the technology is going. I just know that the complexity of the human questions and the human piece is where it can get a bit trickier.
Lorenzo Feruglio:
Yeah, yeah, a hundred percent, hundred percent.
Kelli Kedis Ogborn:
Keeping on this thread then about some of the challenges for expansion. So we talked a little bit about trust. We talked a little bit about these aspects of humans using it in ways that potentially it wasn’t intended, but what do you think are some of the biggest challenges facing artificial intelligence integration for space applications now? And then on the other side, what do you think are the biggest opportunities for its utilization?
Lorenzo Feruglio:
Yeah, so to me, I would say that today the biggest issue in the utilization of AI in the space economy, it’s not really a matter of infrastructure per se, because yes, some missions, some mission control center are quite outdated, and those are not a target, of course, for ai, not on ground, not in flight. But I think to me, what has to be fixed, and that’s the role of our company for example, is letting these companies understand that there are specific use cases in which a lot of value can be obtained by embedding or utilizing AI in your concept. And because today for a given set of reasons, the space economy itself has remained kind of a closed world with respect to the innovation of ai. If you think about the charge GPT and then the language models that have disrupted basically every industry, the flow of these new models and capabilities into the space economy has been really, really limited.
So there’s an educational path and process that needs to happen to make sure that these customer actually understand the benefits they can get by using these technologies and automating their missions. And then once they see the benefits, then getting the technology, it’s a no brainer because you see that you can save $130,000 per satellite per year if you implement predictive maintenance. Now that’s what we see with our product, why you should not getting the technology if this is the result that you can get. But today, most these companies, they don’t know this. That’s the effort of ico, of educating the industry that ICO and many others of educating the industry in adopting such technologies.
Kelli Kedis Ogborn:
Do you think that the slow adoption or the slow matriculation of artificial intelligence into space is purely just the understanding of the efficiency gains that you can get, or do you think it’s potentially just the way that the industry has done business and built systems since its inception?
Lorenzo Feruglio:
Yeah, no, the second aspect is for sure a reality because at the end you need to imagine that these companies have invested millions and millions in building an infrastructure, and therefore those investments are difficult to be changed and to renovated, to be renovated, especially if automating emission really means changing many aspects on how you do operation as a company. So there’s for sure a complexity in really adopting these new technologies. It’s been simplified honestly, because now you can buy a power flow infrastructure already with kind of a not extensive budget, let me put it this way. So it is really ramping up the adoption of AI in many companies that we see. A couple of years ago, many companies wouldn’t even know where to run AI models, and today they started to adopt cloud technologies or they are fine with buying an AI computer. So this is evolving.
Kelli Kedis Ogborn:
Yeah, I think it’s moving definitely in the direction of being, I think less of a buzz word that people know this ecosystem exists to actually sort of, I think, better understanding what it is. I do think there’s still a bit of cloudiness around the complexities, and you said the different language models that go into artificial intelligence. It’s not just one thing, but at least we’re trending in an area where people I think are understanding that it is really critical and there is a necessity to involving it, and I think they’re starting to understand it more.
Lorenzo Feruglio:
I do believe the same. Absolutely.
Kelli Kedis Ogborn:
And then where do you think the biggest opportunities are for utilization? So you mentioned particularly earlier around your company and sort of where it’s evolving, but if you’re looking sort of five to 10 years ahead at the evolution of different types of space mission sets, where do you think we could get the best gains from it?
Lorenzo Feruglio:
Well, this is really from the perspective of an AI company. The most gains in terms of scalability come from being able to support different types of space missions. So for us, it’s not just satellite systems per se, but maybe we can scale. We are already doing it. By the way, working with launcher companies and launcher companies can drive the value or basically the scalability factor because they launched so many satellites and they have so many launches. This is for sure a direction. Another interesting direction is expanding outside of operations towards the earlier phases of the latest phases of the production of a satellite. Now, because overall AI is useful, where there is a lot of data being generated, and when these companies actually do the data, do test the satellite before launching, they generate a lot of data now and they might need some support in there as well. And we see this with our products because we are not supporting just operators, but also companies that are building the satellite and reducing, again, the number of hours spent by the engineers in skimming through the data. These are for sure options. There is always the dreamy theme about deep space economy, the human exploration and the space exploration of other planets or other systems. That’s a appealing direction for sure. Probably it’s less scalable in terms of commercial space. That’s something that we are keeping an eye on as well.
Kelli Kedis Ogborn:
What about the digital twinning space about mission sets and the use of artificial intelligence? Have you put any thought toward that or has your company been monitoring that at all?
Lorenzo Feruglio:
Well, we’re not involved in the overall idea of the digital twins, even though we recognize that the use of AI is actually useful also to basically estimate behavior of a system and then building a more complex digital twin. Our priorities are towards automation today. So as a company, we’re trying to keep the focus on these topics. But yeah, we do see companies working in the field of digital twins embedded AI in their models, so that’s for sure a direction.
Kelli Kedis Ogborn:
Yeah, because that topic beyond the other ones that you mentioned as well, keeps cropping up in conversations just around the necessity to have those sort of systems, especially as you’re looking toward more complex mission sets and environments. And this sort of goes to the deep space piece and a lot of the cis lunar aspects, but I’m starting to see it more in AI is sort of the sister part of that conversation and how it can support that environment. My last question for you, because I think you’ve done a really, really good job, really illuminating how this underpins growth and how these efficiencies and risk mitigation can happen. I’m curious though, I’m trying to think of how to phrase this. What else needs to happen to make sure that artificial intelligence can grow in the ecosystem? So we were talking about ground stations or is it materials? You talked about data centers in space. What could hinder progress that’s not technologically based,
Lorenzo Feruglio:
Aside from the adoption barriers that are maybe those that were, I was speaking earlier, that’s a challenging question, honestly. Well, there are certain topics of regulation, the AI act and some regulation on the use of ai. But as for what we are seeing today, those are typically less striking in the space economy because we typically don’t deal with personal data. It’s really not a domain in which most of these regulation actually do have a strong impact. So it’s a difficult question to answer. Honestly. To me, the adoption barrier is the mentality on these companies on try now these technologies is what we need to solve today, but we are doing it technologically speaking. The scene is evolving now. Satellites are becoming able to execute AI on board, which is definitely something that was needed a couple of years ago, but now it’ss being solved. So I think, yeah, we need to work on really making sure the satellite industry understands the benefits of embedding AI and automation into their frameworks.
Kelli Kedis Ogborn:
And to your point, I think it’s becoming more of an inevitable part of every sort of space conversation that hopefully that challenge becomes a little bit less cumbersome as the months and years tick on.
Lorenzo Feruglio:
Well, you’re seen in our conversation now when you mentioned the concrete KPIs and numbers that you can achieve, it’s really a door opener in the conversation because these companies are starting to realize that they’re missing out on some benefits of ai, so it’s really going in the right direction.
Kelli Kedis Ogborn:
Well, and being able to go from 10 days to two to three hours, I mean, that’s like next level efficiency that I think every company would be striving for.
Lorenzo Feruglio:
And it was funny that I was in a conversation and there was this person, A CEO of a very, very important satellite company that was bragging on how well their platform was automated and so on. Then when there is a failure, they can recover the failure quite quickly. And my argument was really, yeah, but what if you can avoid the failure at all? Wouldn’t this be better than the current situation? Also, it’s important that we keep pushing for the understanding of the benefits of AI in this space.
Kelli Kedis Ogborn:
No, absolutely. And you guys have done such a good job with your company, and I know that it’s continuing to grow, so congratulations on continuing to educate, but also growing in the space. Is there anything that you would want to mention as final thoughts that I did not ask you?
Lorenzo Feruglio:
Not specifically, no. If there is any talent that is listening, please do feel free to contact us. We always eager to hire new talent, so I think that’s my message for you today.
Kelli Kedis Ogborn:
That’s great. Yeah, I imagine that a lot more of the workforce pipeline and the younger generations are going to want to get into this space, so that’s a good plug. Yeah, yeah, absolutely, Lorenzo. Well, thank you so much for joining me today and sharing your insights. I really enjoyed this conversation and I honestly learned a lot. So thank you for that.
Lorenzo Feruglio:
Thank you so much for the invite.
Kelli Kedis Ogborn:
Absolutely, and thank you to all of our viewers. I hope you found this interesting and look out for future vector episodes and please remember there’s a place for everyone in the global space ecosystem. We’ll see you next time. Bye.
Related Articles
Listen to the Podcast
Powering the Future: AI’s Role in Space Innovations and Operations