Experts warn Pentagon to embrace software-defined warfare to counter China’s military advantage

Keyboard with China flag key (Getty images)

WASHINGTON  — If the US wants to win a war against its adversaries, namely China, the Pentagon must get serious about implementing software-defined warfare, experts from the Atlantic Council said Wednesday. 

The Atlantic Council’s new report “Commission on Software-defined Warfare” outlines recommendations for developing a military that pivots from the sole use of legacy hardware and processes to software-defined warfare — a software-centric, hardware-enabled approach that focuses on the continuous integration of cutting-edge, interoperable tech. 

When asked what would happen if the Pentagon did not implement software-defined warfare sooner rather than later, Commission Director of the report Stephen Rodriguez told reporters at a Defense Writers Group event, “we lose to China.” Therefore, he said, the recommendations focus on the “near term” before 2027, when several defense leaders theorize that ​​China will invade Taiwan. 

“China’s outproducing us in ships, munitions and other systems,” Peter Modigliani, one of the authors of the report and a senior advisor at Govini, added. “So that’s where software is going to be the differentiator. Harnessing America’s commercial advantage in a military standpoint, to then have that so we can rapidly upgrade legacy systems, design new weapon systems, and then have the rapid decision support from C2 [command and control] to logistics, and have that rapid iterative cycle. That’s going to be an advantage.” 

Related: To be ready for China in 2027, think new software, not ships: Navy tech officer

Out of the nine recommendations laid out in the report, the authors and other leaders from the Atlantic Council focused on three main ones when talking to reporters: mandating an enterprise data repository and investing in AI enablers; modernizing test and evaluation infrastructure; and enforcing commercial as the “default approach” for software. 

Regarding the mandated enterprise data repository, Whitney McNamara, an author of the report said the platform shouldn’t be “prescriptive,” but it could possibly be related to the already-existing DoD-wide data repository Advana. She said the focus is on collecting as much data as possible and making it “discoverable” across the DoD, something she said is tricky at the moment because of the amount of siloed data. If the DoD can better share this data across the joint force, this will be a critical advantage when facing China, she noted. 

“We arguably have better and more diverse and more mission relevant data than China does because [of] how expansive our operations are, because of our testing and training, our diversity of platforms, our global reach,” McNamara said. “So if we were able to extract that and actually make it discoverable and leverageable across the department, it means that alone could be a meaningful advantage. It’s not sexy or exciting, data collection, but it really is just like a muscle that we need to continue to flex.” 

In terms of modernizing the infrastructure behind software testing and evaluation, the DoD needs to fix the detachment between developers, testers and operators. Instead of these three processes being separate from each other, the people involved at each stage need to better communicate to increase awareness of how a system works, McNamara said. 

Also part of the infrastructure recommendation is to get the technology into the hands of the operators at a faster rate, Christine Fox, a commission co-chair at the Atlantic Council and former acting Deputy Defense Secretary said. Though she acknowledged this idea has been at the heart of innovation discussions within the Pentagon for years, the problem isn’t yet fixed. 

“Don’t get me wrong, you don’t want to just flood the field with stuff that’s not been checked out,” she said. “But as soon as you can, you put the sandbox in the field, and then you get the T&E [testing and evaluation] there, and then it becomes clear which is going to be helpful and relevant.” 

When it comes to enforcing commercial as the “default approach” for software, the report states that “while a preference for commercial products and services is well established in the Title X statute, the DoD continues to develop new software when there are clear commercial alternatives,” adding that “this often results in higher costs, longer schedules, and increased risks.” 

“Commercial software is often updated continuously across a broad customer base, of which the DoD could take advantage. Instead, updating software to address threats and bugs or add functionality takes considerable time and funding.” 

The DoD needs to buy more software from commercial providers, Modigliani said, but it also needs to include vendors at each step of the process, by creating “checkpoints” in the development process.  

He explained that the DoD needs to be asking questions like “did you engage industry?” Adding, “prove to me that you didn’t just do market research and that you have ongoing relationships.” 

Though the report has been in the works for several months, it comes at a convenient time, the authors said. Earlier this month Defense Secretary Pete Hegseth released a memo that called for speeding up software acquisition by directing that the department adopt the Software Acquisition Pathway (SWP) as the preferred method for software development. Hegseth’s memo also directly mentioned software-defined warfare, stating in the first line that the Pentagon “has been slow to recognize that software-defined warfare is not a future construct, but the reality we find ourselves operating in today.” 

The memo directs the use of Commercial Solutions Openings and Other Transactions as the “default solicitation” and award avenue for acquiring capabilities.

Commercial Solutions Openings is a solicitation process developed by the Defense Innovation Unit (DIU) that allows the DoD to work with nontraditional defense companies in a more flexible, faster, collaborative manner. Any contracts awarded as a result of this process are handed out as Other Transaction Agreements.