Lessons from the last shutdown; One-on-one with Bill LaPlante; Gen. Milley talks AI and more.

Yes, I’m sick of writing about potential government shutdowns, but alas, here we go again. Since the fundamental issues creating the gridlock are largely the same, let’s flash back to 2013, the last time the government did in fact shut down.

That 16-day shutdown affected the entire defense enterprise, from the Pentagon to contractors. While the military remained on duty, most Pentagon civilians were initially furloughed. Those who weren’t furloughed included employees at organizations — like U.S. Transportation Command — who are paid using working capital fund coffers.

Days into the shutdown, Lockheed Martin said it would start furloughing some 3,000 employees because the government facilities where they worked were closed — and that more would be idled if the shutdown dragged on. Similarly, United Technologies, then-owner of helicopter maker Sikorsky, planned to furlough 5,000 workers.

Then Defense Secretary Chuck Hagel recalled most of Pentagon civilians a few days into the shutdown, and Lockheed dropped its number of employees impacted to 2,400, most of whom worked in civil facilities.

As well:

  • Flight testing was disrupted for the F-35 Joint Strike Fighter, the Pentagon’s largest program.
  • Weapons production slowed at defense factories because of the absence of Defense Contract Management Agency inspectors.
  • The Pentagon stopped announcing contract awards — although the process went on, and contracts were finally announced after the shutdown ended.

To sum it up, a shutdown is not good for business. Wall Street analysts say defense stocks may avoid a downturn in a shortish shutdown, but the longer it goes, the worse it gets.


You’ve reached the Defense One Global Business Brief by Marcus Weisgerber. Send your tips and feedback to mweisgerber@defenseone.com or @MarcusReports. Check out the Global Business Brief archive here, and tell your friends to subscribe!

From Defense One

North Korea-Bound Ships Could Soon Be Routinely Intercepted by the US and Allies // Steve Mollman

Foreign ministers from the U.S. and about 20 other countries will weigh the measures as part of new pressure on Pyongyang to abandon its nuclear and ballistic missile weapons programs.

Nuclear Command Network Would See Comprehensive Upgrade Under Trump Plan // Marcus Weisgerber

While the ICBMs, bombers, submarines and cruise missiles get lots of attention, the network connecting them also needs updating.

What the Hell Happened in Hawaii? // Ankit Panda

False alarms aren't just terrifying—they're dangerous.

Q&A: Mitre's Bill LaPlante

It’s been more than two years since Bill LaPlante stepped down as the U.S. Air Force’s assistant secretary for acquisition. He works now as senior vice president and general manager of Mitre’s national-security sector — a job that oversees the firm’s 4,000-plus engineers and scientists who support the Pentagon, its federally funded research and development centers that support the military, intelligence community, and law enforcement, and another FFRDC that focuses on cyber security. He’s working to make those FFRDCs respond faster to Pentagon and intelligence community emerging needs — like the new nuclear command-and-control network, automation, artificial intelligence, cyber security and protecting satellites in space and making the security clearance process faster. Here are some excerpts from our conversation at his office in Tysons, Virginia, last week.

Q. What are some of the new things you’re working on and what are some historical projects that might still be relevant today?

A. The thing that’s exploding that we’re working on is autonomy and machine learning and how the marriage of that applies with big data sets. That’s an area that’s across every one of our domains. The second is the modern way that software is developed. The way it’s done in commercial — like in Silicon Valley, you’ll hear the term “devops” and “software factory” — is developed continuously in real time extremely effectively. That has to be brought into the government. Third area is in national security space … [where] the situation remains quite serious. What I mean by serious are the challenges to our superiority. The fourth area is the recapitalization of the nuclear triad, including the command and control. The glue that pulls it all together, which is the command-and-control system is where I’m spending a lot of my time.

Q. That’s the NC3 system, the nuclear network?

A. What does it mean to have a modern NC3 system of systems combined with how decision makers want to make decisions? During the Cold War, NC3 would have been the president on a red phone and a five-minute conversation. In today’s world, it could be the bin Laden picture [when government leaders are monitoring the situation in real-time]. Addressing that is such a fascinating area and it needs a lot of attentional just as you have to modernize the legs of the triad.

Q. On machine learning, is Mitre doing the technology research itself, is it policy recommendations or both?

A. It’s two things. First the technology itself — understanding what are the limits to what it can do. If I wanted to have a machine learning algorithm that can go through massive amounts of video, like YouTube, and find with reasonable false alarm everytime I see a certain individual in any video ... how can I take that technology, combined with say learning technology … [and determine change]. How do I take a Pacific scenario in a modern peer competitor on the surface of the water, under the water and in space, how do I apply that and have revolutionary concepts that help me be superior in that scenario. It sounds like it’s an easy thing to take that make it work. That’s the key place that a place like Mitre focuses. Hypersonics is a fascinating technology. But that technology by itself doesn’t win wars. You have to understand what the concept can do to help us win a fight.To this very day, the United States does not have that as an operational capability because the elements of how it fits into how you use it practically does not work. Another is the airborne laser. It’s a fascinating technology. We have yet to make it an effective missile defense technology. What we want to make sure of with artificial intelligence is that we don’t fall in love with the technology, we — as quickly as we can — understand how it can be put right into warfighting concepts.

Q. Are you working with the intelligence community to use algorithms to help characterize objects in video feeds?

A. One of the things we are doing for the intelligence community is [our laboratories] are a clearing house where the companies that have these different algorithms can come in here and play their algorithms on real datasets. The datasets are surrogates for the real world and they can see how their algorithms are working and we can give them tips. A lot of the key things on these machine learning problems is the to get the best dataset. You can imagine in the intelligence community the datasets are very sensitive. What we do is come up with surrogate datasets that can be openly shared, but that they’re representative of real problems.

Q. Let’s go back to hypersonics and lasers. What needs to happen to get that stuff across the finish line?

A. There is an area of hypersonics for this country that in the last few years, I think and I know others believe, we are potentially falling behind. There are two classes of hypersonics when you talk about strike. The traditional class is: I have a cruise missile with a hypersonic engine on the back, so it’s hypersonic powered. But there’s a second area of hypersonics. It’s what’s called the hypersonic glide vehicle concept, where you have a conventional booster with a glide vehicle on top (which looks like a flying triangle). If you do it right … the simulations say you can go halfway around the world. It’s basically like a skimmer. [Darpa tested this tech in the Pacific several years ago]. Both of them failed. China and Russia are probably further ahead on making that work. If you look at what they’ve done, I would say they’ve progressed beyond us. The two or three times we tried it most recently, where we ran into trouble was in the stability and into the thermal management. The Army had an experiment about five years ago that worked. My impression from what’s in the open press is China has done much more testing and has had more successful flights than we ever had.

Going Faster

One way Mitre is looking to speed up the development timeline is through a project called the  Cross-Cutting Urgent Innovation Cell initiative. It’s acronym is CUIC, but pronounced: “quick,” get it? The corporation is looking to emulate the Pentagon’s rapid acquisition organizations, like the Air Force Rapid Capabilities Office, which LaPlante oversaw as the head of the service’s acquisition projects.

“You find those hard problems where nobody is in charge and don’t let that fact that nobody’s in charge of it stop you from putting it together,” LaPlante said. “That’s a unique thing that I think an FFRDC can do.”

One of the projects Mitre is working on: finding new ways to combat mobile missile launchers, like the ones used by North Korea. The challenge for the Pentagon and IC is these they don’t have much warning before a launch. Among the reason ballistic missile launches are challenging, lots of people, countries, radars, satellites, ships and interceptors are involved.

“Those type of problems are the perfect type of problems for an FFRDC to stitch together,” LaPlante said.

They’re also working on ways to protect critical infrastructure, like power plants to identify common characteristics of potential cyber vulnerabilities.

“We’re trying to drive into people here at Mitre is just because you do a cool demo, doesn’t mean anything,” LaPlante said of the CUIC effort. “It doesn’t mean anything until you get it to a person.”

Gen. Milley on AI

AI is the buzzword of the moment in the Pentagon, and military leaders are already looking for ways to move the technology to the battlefield. “Whether we like it or not, artificial intelligence is coming. Machine learning is coming.” Gen. Mark Milley, the Army chief of staff, said this week at a breakfast sponsored by the Association of the U.S. Army.

Milley acknowledged that experts dispute when and how artificial intelligence will be employed by the military, but said, “Artificial intelligence will happen” and said he’s ready to put money toward it.

“There is a curve to technological development and I don’t know what it’s going to be for artificial intelligence,” he said. “But, I am pretty certain — and I’m willing to bet on it with programs and money and so on — that artificial intelligence is going to have a significant role to play in societies worldwide and in the conduct of warfare.”

Why it matters for the Army: Future vehicles will be optionally manned and the conditions on the ground will determine how they are used in combat.

“I don’t know if artificial intelligence is going to mean robots and machines replace humanity … but I do know that quantum computing and some of the IT technologies that are out there today are so significant and can help you make rapid decision making in complex, decentralized environments that if we don’t take advantage of that and things like the network, then we’d be fools because others are moving out quickly on that,” Milley said.

Upcoming Earnings Reports

  • General Dynamics,  Jan. 24
  • United Technologies, Jan. 24
  • Northrop Grumman, Jan. 25
  • L3 Technologies, Jan. 25
  • Raytheon, Jan. 25
  • Lockheed Martin, Jan. 29
  • Boeing, Jan 31
  • Textron, Jan. 31

Making Moves

  • Melissa Flagg has been named head of the Army Research Lab  — Northeast in Boston, a according to a note from her former employer, the Center for a New American Security.
  • Yosry Barsoum has been named vice president and director of Mitre’s Homeland Security Systems Engineering and Development Institute federally funded research and development center. He was previously MITRE's portfolio director for work with the Office of Secretary of Defense and the Defense Information Systems Agency.
  • Norwegian defense firm Nammo named Stein Erik Nodeland — A retired Royal Norwegian Air Force major general who ran the Norway’s F-35 program — executive vice president for aerospace propulsion.