Frank Konieczny, Chief Technology Officer at the US Air Force (USAF), discusses the importance of remaining mission-focused in an ever-evolving technological world.
How did you find yourself working as CTO for the USAF?
It’s unusual because moving from industry into government is kind of the opposite way that most people do things. I was the Chief Technology Officer, CIO, and Operations Director at a large government telecommunications contractor. One of my friends was at the Air Force and he was trying to get his technology established. I told him I’d go to the Air Force if he could give me a CTO position – a true CTO position.
They wanted an injection of commercial capabilities and commercial experience into the government. It was kind of a new thing. You see it now more than ever, more people from the industry come into government because they want that injection of talent and capability and difference of opinion.
What was meant by a “true CTO” role?
I think I was the first in a position described as CTO in any of the departments, the Army, the Navy, the Coast Guard etc. I was already a CTO and kind of knew what we were going to do, but it was a question of the scope of the responsibilities across the Air Force, which is rather large.
Now we talk about me being a Chief Information Technology Officer because there is another CTO for R&D who covers airplane platform development, such as materials and wingspans. I only get involved in the IT part of it.
The question was: what should a CTO do and what expanse should they have? That’s what is unique about it, they had never thought about doing this before. They had technical advisors come in, but they had never had a true CTO across the Air Force.
What did you bring to the role?
I had both operational experience and CIO experience as well as technology experience, which was unusual. I could look at it from various viewpoints that would normally be more pigeonholed into their viewpoints.
My role was to bring new technology into the organisational structure as it was, which was difficult at first because this was brand new to them and we had to convince everybody this was a good idea. That was a big difference.
Most of the stuff we do in the Air Force involves written requirements. You generate a proposal and then a vendor comes in and wins the bid and everybody’s very happy. But they never think about how you inject technology or what technology you want to really go for because a lot of the stuff was requirements based upon your prior history and your knowledge. I was bringing in new types of capabilities that they hadn’t seen before.
Have you had to work to obtain an “operational buy-in” from multiple stakeholders?
I had to set up what we call the target baseline architecture because that’s the only way they could see something. The difficulty comes from them wanting to see some capability. They want to see something on paper. You have to show them that you have the intelligence and capability to do this.
You also have to present something that says, “Here’s a problem I know you have and here are some technologies that you should start investigating.”
We put this into a target baseline. The target baseline for us is identifying what’s going to happen for a particular problem area within two years; what you should be going for and what technologies are available for two years out.
It was a different way of looking at it because most of the planning cycles for the government are 10 years out. We’re in an age where 10 years out isn’t possible or you can’t even determine as we barely what’s going to happen in the next year.
As an example, early on they were doing some testing on a test platform. When they brought it into the real network, it never worked right. They would say: “I don’t understand. We tested it.” I would tell them that in the industry, you don’t do that.
You test it on a test network and you bring it to the real network and then you test it in the real network and you control it. You just have to make sure because the network is so complex. They never did it like that. They wanted to make sure it was perfect before moving.
I convinced them that they should be testing on the real production network and they should control it in a different way than they were doing it before. A lot of it is a trust issue. They have to be able to trust that you do understand the technology and you do understand some of the problem spaces.
I was a duty contractor before, I had worked with all the components and different problem spaces that they have and I was a project manager for a very large system for the Army. I knew what the problem spaces for the components were.
Do you think you’ve been successful on this front?
It’s taken multiple years to get there, but we are doing different experiments in the way the organisation does things. Everybody believes we have to risk, everybody believes we have to do continuous monitoring, everybody believes that anything we feel is going to have a problem is fixed and has to be more agile.
It’s just a question of introducing people to the capabilities that are really out there and getting them away from the 10-year plan issue.
Have you experienced a situation where organisations look to technology without understanding its true value? And just implement for the sake of implementing?
There are plenty of shiny objects that people want without understanding the ramifications of using that shiny object. Since the Air Forceis distributed across the globe, we get a lot of that coming in the form of “Hey, we want to try this.”
A lot of times they do try and they find out that it’s not extendable to the entire Air Force. I have a rapport with the rest of the organisations, so they come back and say, “Hey, we tried this. This may be interesting. Why don’t you consider it for the entire organisation?”
It’s not a question of stopping innovation in the field. It’s a question of how do you look at innovation in the field and determine if it’s applicable to the entire enterprise and how you would move it to the entire enterprise and support it.
How is the role of the CTO changing?
What you will find in the field, especially in the Air Force, is that we have a lot of officers moving around every two years or so because that’s the normal pattern. They are now depending more upon looking at the CTO as the person that understands the mission and what they need to continue with. That’s the way we established it.
We have CTOs and all the major commands out in the field and a few of the functional commands as well. We have established a foothold, if you will, throughout the organisation, because that’s a dependency. A lot of the officers depend upon the CTO to tell them, “Is this a good idea or not?”
Are you being faced by a new generation of Air Force officer – one that is digitally-led and tech-enabled?
The new officers want everything now. That’s normal. We all do that. Anything new is cool. Then again, it’s just a question of bringing them back into the mission focus, what is really going to support the mission as opposed to anything else.
I think we’ve succeeded in that because we do push them back to the missions, and we’re actually encouraging this now, which is kind of interesting because we have competitions from airmen coming in and saying, “Hey, this is a new technology that we want to produce out there.”
We’re actually giving them money to do things to support their endeavours and everything else. A lot of times that falls back to me or one of the other CTOs to actually watch them to make sure they’re doing it correctly.
We also have more competition now from small businesses. We actually support research and development of small businesses to put new technology out into the field and we actually work with people. So, we have turned ourselves into a technology engine looking at various technologies rather than just writing RFPs and sending them out.
What’s changed about your role since joining the organisation?
Most CTOs don’t have airplane manufacturing associated with them. In industry, most CTOs are CTOs basically for IT. The only difference is we change the duty title a little bit because we wanted to emphasise that we’re focusing on IT as opposed to anything else, but as opposed to actually doing material testing for new wings and evaluating the capabilities and the vibrations of wings, giving new designs and everything else and the engines associated with it, things like that.
We have this way of looking at the R&D effort before it gets into a technology point where you can actually feel it. That’s the R&D piece of it. The IT piece is everything after that, basically.
When you talk to the CTO of most organisations they will tell you they’re all IT because there may be some pieces of manufacturing, but most of them are not in the manufacturing area except for watching, making sure the equipment actually works. That’s not what they’re doing.
They’re not designing manufacturing equipment per se, unless again, there’s an IT component of automation, artificial intelligence and everything else. It’s not that I don’t work with things on airplanes. I just don’t design airplanes.
How do you ensure that, as a technology professional, you are continuing to learn and remain ahead of the game?
I read a lot. That’s nothing new, but I try to stay abreast of what’s going on. Our tech vendors keep me abreast of what’s really going on with their push. I look at what technology is actually happening and where we should be going. It’s just one of those things.
I read a lot in the field and see what’s happening so I can see where we’re progressing, and the question then becomes, “How can we best move in that direction?”
What are some of your current initiatives as CTO?
We’ve been trying to bring mobility to the airman for a while now. The aircraft are mobile but for the airmen, mobility is a big key because what we’re trying to add connectivity to an iPad or whatsoever and send information on it to them because there’s no connectivity out in the field, on the airfield. One of the big pushes right now is pulling LTE and 5G out into the bases to start doing some of this capability.
Does it form a broader question as to what are we ultimately looking to do? We’re trying to make sure that aircraft get maintained quickly, effectively, in a certain way. We want to make sure that the parts that we want can be ordered directly right there when the repairman is there trying to fix the aircraft.
It’d be nice if we were sitting at a depot and the part could automatically be automated, being moved out to where the repairman is in some automated vehicle. That’s one of the ways we look at going forward with the mission because that mission is important.
Where do you think the next industry shift will come from?
The next shift is AI. It’s such a buzzword right now, but we’re starting to see more and more of augmented support via computers, via neural nets and machine learning capabilities. We’re seeing it more and more and we’re seeing some places where we can actually start using it now. I think that the push is how to effectively use AI technologies to enable a mission. Because a lot of people look at it and say they need it, without even knowing what it is.
With new technologies and faster processing speeds, we can achieve more results with AI. As the processing has increased capability, we see that there’s some applicability for AI to run in real-time. We’ve been doing AI for around 20 years. I was coding expert systems way back when they were just coming out. Right now, we have processing capabilities that support some of this.
At the USAF, we’re trying to analyse data and be more of a data-driven organization, so long as it supports the mission. Everybody says, “Hey, I have to have it,” and you’re like, “Great. Give me a problem and I can tell you what applicable AI techniques there are for that problem space”. As we progress, you’re seeing more and more of that occurring, even though it’s still hype.
How important will people remain?
People believe that their job is going to go away because of AI. AI is just an enabler to do your job better. You still have to be there. We talk about autonomous operations such as autonomous cars. Autonomous cars have a lot of problems with them right now because they have to make decisions really fast and they have to do it correctly.
Then there’s the ethical behaviour of automated entities. We’re going through this right now with AI. AI is just code somebody coded in a particular way. There may be some bias in that code as to conclusions, but you don’t know that. So, you have to understand all this. You just can’t say, “Hey, it’s really great. We’re going to go forward with it and proceed,” because that’s not how it works.
Everybody has to realise, with new technologies that it’s difficult at times to get people grounded into the mission that the new technologies are supposed to support. You’re solving a problem with these new technologies or you’re helping to solve a problem, but this problem is basically something that you want to enhance in your mission. You have to think of technology as an enabler for your particular mission. A lot of people forget that. They just think, “I want to have new technology because it’s cool.”