Q&A With: IARPA Director Lisa Porter
The first director of the Intelligence Advanced Research Projects Activity discusses the differences between intelligence work and defense
As the new director of the Intelligence Advanced Research Projects Activity, Lisa Porter is the United States' answer to James Bond's Agent Q, but she's not crazy about the label. Porter is not the kind of person who likes being reduced to an easy metaphor, nor does she want her agency's intelligence work reduced to easy metaphors. That makes her the perfect head of the new agency, which has been tasked with developing technologies so far out that not even the Defense Department would fund them. ”We're not interested in the near-term, the low-hanging fruit,” she says. Porter wants the tough problems, a characteristic that's reflected in her eyebrow-raising résumé: She received her Ph.D. in Applied Physics from Stanford University and then spent some time as a program manager at the Defense Advanced Research Projects Agency's Advanced Technology Office. She left DARPA to manage aeronautics research at NASA. In January of this year, intelligence director Mike McConnell plucked her from her NASA post to lead the new intelligence agency.
IARPA (pronounced EYE-arpa) was created after the September 11 attacks as part of a larger effort to get the far-flung elements of the U.S. intelligence community to talk to each other. The new agency will be a high-risk research crucible for the country's 16 intelligence agencies, and not just the big ones that everyone knows about (CIA, NSA): many parts of the federal government, including the Department of the Treasury, have their own specific intelligence offices (for tracking money and counterfeiting, for example).
Later this week, IARPA will announce its split into three program offices, which Porter says span the scope of the intelligence problem: Smart Collection, Incisive Analysis, and Safe and Secure Operations. Porter would not get specific about the projects IARPA will work on because most of those projects will be classified.The agency's offices are in a fenced and guarded National Security Agency compound at the College Park campus of the University of Maryland. By next year, however, the agency plans to make its home in a much more accessible part of town, in part to collaborate more with the academics among which it is nestled. IARPA has a lot of pretty big spots to fill: Porter needs directors and program managers for the three new offices.
IEEE Spectrum's Sally Adee talked to Porter about the future of IARPA, the details of the new programs, and what exactly the difference is between intelligence and defense.
IEEE Spectrum: IARPA is tasked with high-risk, high-payoff advanced intelligence research. But doesn't DARPA already cover this? Is IARPA's mission redundant with any of DARPA's programs (like the former Information Exploitation Office?
Lisa Porter: No. It's important that the intelligence community has a place to focus on its own kind of high-risk, high-payoff research. When we talk about high-risk, high-payoff research, we're not talking about low-hanging fruit. This is about the really hard problems--we have good ideas, we may not succeed, and that's completely acceptable. That's the same realm DARPA operates in, but not specifically for intelligence.
Sometimes DARPA creates dual-use technologies that have a defense purpose and an intelligence purpose, but in that case the intelligence purpose is often incidental. It's where there is crossover between the defense mission and the intelligence mission. In that case they'd partner with intel agencies--but they are focused on the DOD mission and not the intelligence mission.
Spectrum: Can you explain that? People tend to conflate intelligence and defense. They seem to both be about defending the country.
LP: They're conflated because they're partners and they often work together. Sometimes--many times--they work together, but the intelligence community provides strategic information so that decision makers can do what they need to do. There are times when they work together to accomplish something and also to invest in dual-use technologies. But there are differences in the two missions. For example, intelligence gathering has different timescales. Think about the 16 intelligence agencies: you have the CIA's mission, the NGA [National Geospatial-Intelligence Agency], the NRO [National Reconnaissance Office], the NSA, and they are all collecting, assessing, and analyzing intelligence information, and they're not necessarily focused on a near-term tactical tempo or defense tempo, or even defense applications.
Spectrum: So how does IARPA work? What problems are you looking at?
LP: We've divided the agency into three offices, and those really explain how we parse the problems intelligence research is focused on: Smart Collection, Incisive Analysis, and Safe and Secure Operations. Those three thrust areas span the space of the intelligence problem.
The first, Smart Collection: we want to dramatically improve the value of our collected data. It's not enough to collect data. You want to do it smart, because you're often limited in the amount that you can collect. It's the classic problem of the drunk looking for his keys under the lamppost--not because that's where he dropped them but because that's where the light is. You fall into that trap a lot. You look where you know how to get, not necessarily where you need to be. So we're trying to use modeling and analysis to help us look elsewhere than where the light is.
The second office is called Incisive Analysis, where we look at maximizing the insight we get from collections in a timely fashion. Analysts are drowning in reams and reams of data. It's called the tsunami effect--the overwhelming amount of data and information that they have to analyze. How can they go through it all fast enough to provide decision makers with analysis in time? There's so much information out there--I mean, just go look at YouTube. Think about the information in your life--all the e-mails that you don't have time to read. In this office, we're hoping to get smarter about data analysis, maybe by using virtual worlds. How do we leverage some of the creative ideas that might be out there to help our analysts get their arms around all this data? And there's another, multidisciplinary aspect of this problem: ideally, you want to understand not just what's being said but the cultural implications as well.
The third office is called Safe and Secure Operations. Here we want to counter the capabilities of our adversaries that could threaten our ability to operate effectively in the networked world. That includes the challenge of cybersecurity. When you talk about security, you're talking about confidentiality, integrity, and availability of the system. This office will also work research into quantum information theory. That's a very high-risk area.
Spectrum: DARPA programs are often transitioned out to civilian use: the Internet, GPS, advanced prosthetics. Are you looking at transition partners for civilian applications?
LP: Sometimes. When you come up with a new idea and you do the prototype, sometimes your transition partner is a commercial partner. A new software capability, for example, can sometimes go right to a commercial vendor. There are also examples where the commercial sector could also be the beneficiary. DARPA has often advanced technology in a way that naturally benefits the private sector: the Internet is your classic example of that.
I anticipate that since the problems we'll be addressing are very hard, we'll be advancing technology capabilities, and that will spill over into commercial or private-sector applications. That happens a lot with cutting-edge research. You do often see applications that you didn't even anticipate.
Spectrum: Are there intelligence applications that do not overlap with defense purposes?
LP: Yes, there are.
Spectrum: Can you give me an example?
LP: I'm sorry, the first examples that come to my mind are classified.
Spectrum: Can you think of any analogues in the nonclassified realm?
LP: There are things the intelligence community has to do that Defense can't justify spending its money on. You know, there are different pots of money, and the Defense Department has to make sure that what it spends money on applies to a defense mission....Some things cannot be justified being related to a defense application; it's really something the intelligence people have to take on.
Spectrum: Why is IARPA at the University of Maryland, College Park?
LP: That decision was made before I came on board, so I might not be the best person to answer why we're here. It makes a lot of sense in terms of proximity to a university, proximity to the D.C. area. It's easy for us to get anywhere we need to get; we can get to our various intel agencies. We want to look across the agencies and across the community; we don't want it to look like we're only here as one specific agency. It's nice not to be sitting right next to one particular agency. It's also nice to be near a university because we're sending a message that we want to bring in nontraditional partners: academia, industry. It sends a nice message that we're embracing the broad community to help us solve these challenging problems.
Spectrum: So they'll be able to get past the black gate?
LP: Exactly. We want to send that message: we really want to be outward looking and engaged in the community. We're here for everyone, looking across agencies' problems, and we're academic friendly, though obviously still friendly to people who are used to working with us. We're open to people who may have thought there was a barrier to the intelligence community in the past.
Spectrum: Is there anything from DARPA that you are bringing with you to IARPA?
LP: Yes. A lot: the way DARPA approaches, and is really true to, the high-risk, high-payoff thing, for example. It's really important not just to say, ”I want to solve this hard problem.” You have to have an idea to solve it, and you have to have a good program manager to lead it. [DARPA director] Tony Tether has said many times: DARPA will not start a program without a good idea and a good program manager to lead it into reality. It's not enough to have a good idea. It's very hard to be a program manager.
And the flexibility. I was in the Advanced Technology Office, an office that no longer exists. That's a hallmark of DARPA, and it testifies to the flexibility and fluidity of ARPAs.
Spectrum: Will any DARPA projects be transitioned into IARPA?
LP: Not right now. I'd say there are areas where we'll work together. I see opportunities for IARPA and DARPA to work together, but frankly I'm still pulling things together, and identifying people I may want to bring in.
Spectrum: Are you working with In-Q-Tel?*
LP: I've met with them. They're working on high-risk but near-term stuff. They're very impressive. We want to make sure they're aware of what we're doing. I think we can complement each other.
Spectrum: You're looking at getting program managers. But even DARPA has trouble finding those. What's your strategy?
LP: Well, part of it is to talk to people like you to help get the word out. We're looking for very smart people who understand what it takes not just to technically comprehend a problem but how to bring an idea to reality programmatically. It's not easy.
What we offer is, okay, so you have this great idea. If you can convince me, we'll give you the opportunity to make that idea a reality. You can take a risk--and failure is okay. This is a great place for people with a great idea. It's really risky, the potential payoff is huge, and failure is okay--that kind of environment is pretty hard to find.
The IARPA.gov Web site will be up by the end of the month, and we'll have information on how to apply. It requires good tech expertise, programmatic knowledge and expertise, and a willingness to put all of your passion into it.
Spectrum: Kind of like a Make-a-Wish foundation for geeks.
LP: Yes, it is. And I need three good leaders for the program offices.
*In-Q-Telis an Arlington, Va.–based nonprofit private venture capital firm that invests in high-tech companies (in three broad technology categories: software, infrastructure, and materials sciences) to support the CIA's information technology needs. Established in 1999, In-Q-Tel bills itself as the”skunk works” of the intelligence community.