The U.S. military’s new F-35 stealth fighter is again falling behind schedule in its 16-year, $60 billion development. The problem this time—the radar-evading plane’s 8 million lines of computer code, amounting to arguably the most complex software suite ever installed on a warplane.
The code delay is the latest—and possibly most damaging—setback for the Pentagon’s ambitious and controversial plan to replace almost all of its Air Force, Navy and Marine Corps fighters with three different versions of the F-35 at a cost of more than a trillion dollars over the next 50 years.
Damaging, because the military and F-35-maker Lockheed Martin have increasingly sold the F-35 as a sort of “flying computer” whose software can outthink enemy pilots even when the enemy’s own planes fly faster, maneuver better and carry more weaponry than the F-35 does.
The stealth fighter’s software is its last possible claim to being a first-class warplane. If the F-35’s code doesn’t work, then neither does the F-35. Saddled with thousands of dysfunctional F-35s, the Pentagon could lose command of the air.
J. Michael Gilmore, the Defense Department’s top weapons tester, warned about the schedule slip in a Dec. 11 memo. “The current ‘official schedule’ to complete full development and testing of all Block 3F capabilities by 31 July 2017, is not realistic,” Carter wrote, referring to the software update, or “block,” that’s supposed to give the F-35 the basic ability to use its sensors and some weapons—albeit with limitations.
According to Gilmore, the Block 3F code delay is a consequence of the F-35 developers’ rush to install the earlier Block 2B software, which is suitable only for testing and training but is supposed to form the basis of the later, combat-ready code. The developers’ hurry with Block 2B resulted in “poor performance” that slowed progress on subsequent code.
Realistically, there are only two ways to fix the Block 3F software, Gilmore noted. One is to triple the pace of testing. But caving to budget cuts and the enormous—some might say overwhelming—challenge of developing a jack-of-all-trades stealth fighter, Lockheed and the military are actually planning on slowing testing by two-thirds.
The second possible solution is to strip some features from the Block 3F software—say, compatibility with certain high-tech weapons—and wait to add those capabilities back to the F-35 on later software blocks.
But either option means a “very high risk of failing” when the F-35 with Block 3F code undergoes its developmental final exam in 2018, Gilmore warned. In a move with more public relations value than military utility, in July the Marine Corps declared a single squadron of F-35s combat-ready with the Block 2B software. The Air Force plans to declare a squadron combat-ready in late 2016 with an interim software called Block 3I, but the flying branch is waiting until the Block 3F code is ready before it clears its F-35 squadrons to deploy to the most dangerous conflict zones.
And rushing the Block 3F software could have the same deleterious effect on the next batch of code that rushing the Block 2B software had on Block 3F. If the F-35 developers can’t get the current software update right, they risk derailing the stealth fighter’s entire development. It’s no exaggeration to say that the future of U.S. air power rests on these 8 million lines of code.
There’s a third potential fix that Gimore didn’t mention in his memo. The Pentagon could preserve the Block 3F’s features, sustain or increase the pace of testing and delay the final exam in order to give the military’s coders enough time to finish the software, and finish it right.
The problem is Congress, which in 2012 pressured the Air Force to have some F-35s ready for combat no later than December 2016. Air Force leaders had wanted to wait until 2018 at the earliest for any declaration but were wary of “the negative connotation” of “being two years behind the Marines” with Congress breathing down their necks, according to the flying branch’s official history for 2013. So the generals approved the current schedule with its interim software block and hard 2018 deadline for the final exam.
Delaying the F-35’s war-readiness would look bad for the Air Force. But failing the final exam—or sending unready F-35s into combat—would look even worse.