The Defense Department needs to set up a “parallel” acquisition system built around digital methods and speeds, alongside the current system set up to manage “Newtonian” weapons development, in order to achieve deterrence, Lockheed Martin CEO Jim Taiclet said this week.
Speaking at the Hudson Institute on Oct. 4, Taiclet said the pace of technology development—particularly software and data distribution—continues to accelerate, and the Pentagon needs an acquisition system that can keep up with “a clock speed that is much closer to the digital technology development clock speed” compared to the world of ships, jets, and satellites that operate physically, rather than digitally.
“I think that the U.S. defense enterprise is still the most effective in the world. I think we can deter conflict effectively today,” Taiclet said. However, a separate system for working with digital is “one thing I do think we need to do to stay ahead of the evolving threat.”
The Pentagon needs to shift toward a more commercial-like pace of development measured not in years, but weeks and months, Taiclet said, asserting the answer likely lies in buying digital services, something the Pentagon and Congress are ill-configured to do and are taking too long to accept in his mind.
Simply put, the “general procurement system” is not well matched to digital products, Taiclet said.
“The digital technology cycle is, again, months or weeks instead of years and years. So we’re suggesting to our government customers to think about the procurement and acquisition process differently for digital technology insertion versus Newtonian platform production,” Taiclet said. “That hasn’t caught on yet. And I think it’s something we really need to advocate for.”
The existing acquisition system involves setting requirements, requests for information and proposals, and competitions that can years, if not decades.
“It’s fine on the bigger physical technology items. It does work. It’s been successful,” Taiclet said.
But with digital systems—particularly large data-sharing networks, technology moves too fast so that by the time Pentagon buys software now, it is outdated.
“So that’s the notion of the parallel path, which is, we’ve got to be able to deal with these companies that generally work off of a subscription model,” Taiclet said. “ … Just your cell phone service for example, you pay every month for it. They continually upgrade the network, you’re getting new features. Another app comes on your phone, the app gets upgraded every night, and this is continuously happening.”
The Defense Department doesn’t have an effective way to acquire services like that, he said, and the longer that takes to appear, the more behind it will get.
Taiclet suggested the Defense Department might also buy mission capability as a service, “by the month, by the year. We have to figure out how to translate the DoD form DD 250” which is the process by which the Pentagon accepts final-version goods, “into a subscription service, so I can use Verizon 5G algorithms.”
Lockheed, he said, has teamed with Verizon, INVIDIA, and IBM Red Hat “on managing [artificial intelligence] digitization through a network. We’re partnered with Intel on … chip design to make sure that we can get our requirements into the next venture production line.”
“We have to collaborate with these companies, which our industry isn’t typically used to doing, and the government is not used to paying for,” he said. The Pentagon also has to make defense contracting attractive, with adequate margin so those firms and small startups don’t simply focus on the far more profitable commercial market.
The commercial market also bears lessons for the Pentagon—the telecommunications industry, for example, when through a period with three sets of standards, developed by Nextel, Quaalcom, and GSM. The resulting networks were “expensive, inefficient and incompatible,” said Taiclet, who previously served as CEO of American Tower, a telecommunications infrastructure company.
Taiclet said he’s like “to skip that stage and go right to the single standard, which is what we have now. It’s called LTE: Long-Term Evolution for 4G.”
Along with with a parallel procurement process, Taiclet said, the defense enterprise should “also establish a standards body like we have in telecom … to basically get commercial industry, aerospace industry, government customer and the investor and startup and new entrants together to create a standard that we’ll all compete on and all develop together.”
Accelerating the pace of change in the Pentagon is critical, Taiclet suggested, to deterring a potential adversary like China.
Citing the principles of Sun Tsu, Taiclet said China is biding its time, waiting for “90 percent certainty or expectation of your success.” To deter that, the U.S. and its allies have to constantly “move that 90 percent goalpost” he said.
Therefore, the Pentagon, along with its allies and industry partners, needs to constant move that goalpost—“not every 10 years when we can build a new airplane or a new Aegis radar,” Taiclet said, “but every three to six months; how do we help DOD and our allies move those deterrence goalposts every three to six months?”