For years he’s been at the vanguard of examining emerging technology. And whether it can benefit the federal government and its operations. He’s also delved deeply into the practical applications of data, analytics, and the efficient deployment of technology. Now, he’s a new inductee into the National Academy of Public Administration. Dr. Tim Persons is Chief Scientist and Managing Director of the Science, Technology Assessment, and Analytics Team at the Government Accountability Office, and he joined Federal Drive with Tom Temin.
Tom Temin: And Tim, that is an awful long litany of things. Maybe we should just call you excellency of technology.
Insight by Carahsoft: This exclusive e-book demonstrates just how far agencies have come and where they still need to go to take fully advantage of DevSecOps to drive modern capabilities to their customers.
Tim Persons: I think that’s overkill, Tom, but thank you for the kind introduction. It’s great to be here.
Tom Temin: And congratulations on being inducted. And I want to start with really after looking at this for so many years and following what’s going on with the DATA Act and related developments. How deeply inculcated is data analytics based decision making, do you think, really in the federal government at this point?
Tim Persons: Yeah, thanks, Tom. It’s a great question. Because they’ll sound like a GAO report, meaning progress is made, but more could be done. And I think that we’re seeing trends where data is being used more. I will just say historically, that all federal agencies operate on data, it’s just the nature of the extent of the data, how much, have a thought about mixing data together in certain ways. And we are in an era of the algorithm and high speed networking everywhere, and all kinds of computing resources. So it really has opened up, I think, the aperture on the ability to try and move from a previous computational type framework into one where you can now do a whole lot more, answer much more complex and complicated questions. And the arrival of that is none too soon. So there’s a lot of work still to be done. I think there is still an underestimation generally about the power of data and what that can mean. I think we’re seeing the power come to, we’re waking up to it nationally, as you know, from things like the January 6 incident, or things like how eerily accurate the E-commerce bots can be in suggesting your next purchase. But I think we’re waking up to it, it’s some challenges, but I think there’s some opportunities in the federal government, I think, is ideally situated to have a bright future in this area, if they take the right path and move in that direction.
Tom Temin: You know, at one time, data analytics was simply trying to generate a report out of your Oracle database based on your application, and that had to go to the IT department and then six months later, you might get the program to give that report. Nowadays, there seems to be a nascent democratization, if you will, of the tools — artificial intelligence tools, automation tools — that can be actually put in regular line people’s hands. Again, I don’t think it’s all that widespread. But that’s what the industry is promising. In your assessment, does that have promise?
Tim Persons: I think it absolutely has promised. With DARPA, we think about AI in terms of waves, right? The first wave is sort of like the TurboTax-ey wave where you have the rules are set up front, you can ask questions, and it’ll take you down pathways. We’re seeing a lot of promise in success already with things like lower end AI like robotics process automation, which doesn’t sound like it’s very fancy. TurboTax is a here and now kind of technology. But it saves a lot of time and a lot of cost and makes lives easier for a number of feds who are using it day in and day out. And so the taxpayer’s the winner with that, and that’s just at the phase one. So now when we move into other things — more advanced AI — that’s a scarier proposition. It’s more statistical, we want to have certainty in all things. But nonetheless, there’s a lot of promise in in usage of statistical AI or wave 2. And so that’s where I think there’s a lot of movement to do that. And there are applications from Defense, all the way into, you name your civilian sector issue area, and we can see AI in the future doing that.
Tom Temin: And the other piece of this whole equation, of course, is the data itself. And recently I was talking with Taka Ariga, who was your colleague at GAO, the chief data officer, pretty sharp cookie, you might say too, but the CDO function again, that’s something GAO has watched to see if that is developing in the federal government and most agencies now have the CDO, that is congressionally required, how effective, where do they stand in terms of potential scale at this point, the chief data officer function, would you say?
Tim Persons: So I’m excited about that movement in statute. And then we’re seeing the agencies do it. Some are capitalizing on their data more than others. Sometimes it’s a rate of evolution. Sometimes it’s the system that they operate in and so on. But I’m excited about the CDOs that I see around federal government inclusive of for us, Taka, and it was great to bring him on, because we’re working to help just transform how GAO provides really knowledge and wisdom to the Congress. We have a lot of data but how we convert that from questions into answers, and do it in a more incisive and deep and even quicker, faster way at a time where all the challenges are coming upon us so quickly. It’s exciting, but, I think there are still some underappreciation for data. I see times where maybe the chief information officer, another key role, sometimes that philosophy is almost a hindrance, because you view data as a burden, I’ve got to secure it, I’ve got to lock it away, right? That kind of thing. And so you need a natural and healthy tension between the CDO and the CIO. And they both are equally important in their own right. And you have to win with both. But I think that where I’m seeing agencies succeed is when you have a strong C-suite presence of the CDO helping capitalize on data or start to shift the mission space toward data driven type decision making.
Tom Temin: We’re speaking with Dr. Tim Persons. He’s chief scientist and managing director of the Science, Technology Assessment, and Analytics team at the Government Accountability Office, and now newly a fellow at the National Academy of Public Administration. And that technology assessment title, we hear that term a lot, and GAO has always had a technology assessment team — it was an office at one time in the White House many years ago — what is technology assessment exactly? It’s more than just reverse vendor days.
Tim Persons: Right. So it’s more than just explaining or giving a one on one on science and tech, that certainly is important in terms of explaining to our lawmakers, it really is a foresight type function that’s trying to say, look, given the advent or the emergence of this technology and what we see it doing, let’s project forward and say, what might this mean? And really explore what always exists with technologies Tom, which is a double edged sword. There’s an upside and a downside to every technology without fail. So we have to figure out like, what is that upside? What are the downsides? What are the economic, ethical, legal, social implications of that? So it’s not just a technocratic reporting, it’s like a holistic view on what this might mean. And especially for where we sit in GAO, what this means to policy. So we want to try and forecast not forecast in a, trying to predict the future, but just trying to say if this occurs in this way, Congress may need to think about these particular things in terms of using it. There’s going to be a lot of energy to try and instantiate the technology. But you have to say, well, what laws does that threaten potentially? Or do we have to change things or what’s prohibiting things? And so I think that’s where really the technology assessment comes in to try and give that narrative.
Tom Temin: Because that could be difficult. I was talking to Dr. Collins, about to leave the directorship of NIH, and I asked him how are we doing on genomics? I mean, we mapped the thing 20 years ago. And he said a big issue with that type of technology is it’s often over sold in the near term, undersold in the long term. And there is not much really progress short term, but no one thinks other than it’s going to be a profound change long term. So that’s kind of the thing you’re talking about.
Tim Persons: I think so. And then the tying it in with the data and analytics, because genomic sequencing is just creating data just to pertains to biological issues. And that kind of genomic sequencing has been critical or central to the record breaking advent of our vaccines in the middle of the pandemic, which we’ve never seen before. But I think the next era of genomics is like in that data analytics. We can sequence a lot of things very cheaply now. But the key question is, what does that mean? What sequences this snippet mean, in terms of this disease or that condition? And how might we treat it and things like that? And when you pull in, like the CRISPR, this is the gene editing technology. Now you’re getting into this whole thing about are we playing God? What are the ethical issues? There’s, of course, it’s abundant with that. And yet, how do we hear things like Alzheimer’s or cancer or whatever, in the life science space? So it’s an exciting future, but we have to think differently, and we have to kind of fuse disciplines like computational systems, as well as biologics. And I think that’s going to be the way forward.
Tom Temin: Well I hope they map me into being JS Bach. That’s what I would have been in another life, that kind of talent if I had it. But the other challenge for technology assessors, too, is picking what to assess. For example, the transistor was made in whenever it was made, 1959 or something, nobody, I don’t think could have anticipated integrated circuit with 2 million transistors packed into a thumbnail. Facebook, a small little network of people, now it’s turning into in some sense, this great power, in some sense, a great malignancy. So how do you know what to pick to assess to say this could have big impact?
Tim Persons: Right. So what we do is we look at emerging trends, we do a lot of horizon scanning kind of work. So that’s one area. And then of course, GAO is keen on all of the congressional committees in their issue areas, ranging from again, social security issues, education issues, to nuclear waste management. So we can look at the sort of technological trends in any of those areas and say, what might this mean toward that mission? You know, I really liked your example of the transistor. Because the first application wasn’t a computer, it was really for the hearing aid. And the design principle was I have to have some device that is small enough to fit inside the human ear. And nobody ever thought of the transistor as like, exactly as you said, like, we have computing everywhere. We can do all these sort of things. And it’s associated with that and rightly so. But that wasn’t the original. And that’s the harder part about, no one is good at guessing that, we’re not going to ever claim that we will do it, but we will do our best to try and I think just project forward, lay out or tell the story about what these things might mean and then adapt and adjust as things come about that will surprise us. And we will see surprises.
Tom Temin: And earlier in your career, you were in the intelligence community for a brief period of time. What did you learn there that’s informed the rest of it that you can tell us? Oh, sure.
Tim Persons: Oh sure. Well, a lot of what I was working on, I was the chief tech officer for the Intelligence ARPA, so the Advanced Research Projects Activity, great mission and did a lot in what used to be called the cool chic thing was big data, and so on. So we were doing that at the time. And that really taught me the power of what it means to amalgamate and to be able to see data as an asset and think creatively about it and how people like Taka and his team, come across it and how do we better answer questions to, or convert questions to answers. So I think that was part of it. I also dealt with disruptive technologies, Tom, like quantum computing and other things which are profoundly important. We just actually issued last week a tech assessment on quantum computing. And we really wanted focus on the commercial side. We know it’s a threat in terms of defense systems. And we’re concerned about that. On the other hand, we want to be able to say that we might be able to model the next drug discovery operation, we might be able to solve equations for FedEx or whatever better with that technology. So I learned that really, we are the top innovation country, keeping our innovation capitalizing on it, finding that double edged sword, being agile and so on is I think the open question. But I think we could do it, we have to think differently about it.
Want to stay up to date with the latest federal news and information from all your devices? Download the revamped Federal News Network app
Tom Temin: Now unlike a lot of NAPA fellows inductees, you still have a day job at GAO, and a pretty all encompassing one. What will you focus on for NAPA?
Tim Persons: So for NAPA, I’m really excited about, they have a project on Agile government, and my team actually did an Agile best practices guides. And Agile was always thought of in software development. That’s how, if you have your smartphone, you can update apps, probably every week or so they’ll update to some new version. Agile is what’s allowing you’re driving those sort of iterative solutions. So I think what excites me is that using Agile methods with the fusion of data and analytics, and really overall, Tom, it’s a macro transition of the federal government into a digital services type framework. The government is all about a lot of services. But how do we use digital to really transform and do that for the good of the taxpayer? How do we do it 10x better or 10x cheaper or 100x or so on? And so I think that’s what I’m excited about. NAPA is this Agile driven transformation into digital services across the federal sector. So that’s inclusive of AI, inclusive to cybersecurity, data analytics, you name it. I think there’s a bright future for that if we do it right.
Tom Temin: All right. Dr. Tim Persons is chief scientist at the Government Accountability Office, and a new inductee to the National Academy of Public Administration –NAPA. Thanks so much for joining me.
Tim Persons: Thanks, Tom for having me. Great to see you.