In March 2025, the Pentagon made a decision that got far less attention than it deserved. The Department of Defense told its organizations to treat the Software Acquisition Pathway as the default way to buy software, not one option among many, but the preferred route. On paper it reads like procurement housekeeping. It isn’t. It’s an admission the defense industry has been circling for years: software has stopped being a feature bolted onto hardware. In a growing number of systems, it is becoming the system.

The Pentagon isn’t reacting to software itself. It’s reacting to the speed at which warfare now evolves. Procurement is simply where that realization became official.

Here’s what that shift actually means. For most of the last century, deployment was the finish line. You designed a system, tested it, built it, delivered it ,and then left it more or less alone until a modernization program came around a decade later to rip it open and start again. That cadence worked when the threat environment moved at the pace of hardware. It doesn’t anymore.

Electronic warfare shifts week to week. Drone tactics that dominate one month get countered the next. AI models improve continuously, and operators generate new battlefield data every day. A platform that cannot absorb those lessons after deployment starts falling behind almost immediately ,not because its hardware failed, but because its software stopped learning.

None of this makes hardware less important. It makes software far more valuable. The physical platform is the foundation; the software decides how long that foundation stays relevant. And more and more of what actually wins autonomous navigation, target recognition, mission planning, electronic warfare libraries, sensor fusion, swarming behaviors, and decision support lives in code, not in the airframe or the sensor. The platform stops being a finished product and becomes something closer to a living system.

We’ve watched this change the questions we ask.

A few years ago, technical due diligence at UAX was mostly about the hardware. Does it work? How far does it fly? What can it carry? How accurate is the sensor? Those questions still matter you can’t software your way out of a bad airframe. But on their own they no longer tell us much about whether a company will still be competitive in three years.

Now we push on a different set. How fast can they ship a software update and have they actually done it under real operational load, not just in a demo? Can they add a capability without redesigning the platform? Is the architecture modular enough to absorb technologies that don’t exist yet? When operators send feedback from the field, does it turn into a release in weeks, or vanish into a roadmap for eighteen months? Is there an engineering organization built for continuous deployment, or one that still thinks in periodic upgrades?

Those answers usually tell us more about the next five years than any spec sheet.

The same pattern is becoming visible across the broader defense ecosystem. Companies like Anduril, Shield AI, Helsing, and Palantir have all built significant competitive advantages around continuously improving operational capability after deployment rather than treating software as something that gets updated every few years.

We’ve seen the same trend in our own work. Of the last 15 companies we visited in Israel that were actively raising capital, 13 were building their competitive advantage primarily in software rather than hardware. That’s not a coincidence. Startups rarely beat the primes by building bigger platforms. They beat them by making platforms smarter, more autonomous, more adaptable, and capable of improving long after deployment.

This is the same direction the Pentagon is pushing. Its Software Modernization Strategy emphasizes continuous delivery, software factories, DevSecOps, and delivering capability “at the speed of relevance” instead of relying on traditional multi-year acquisition cycles. The objective is no longer simply to deliver software. It’s to continuously improve operational capability after deployment and that’s a fundamentally different objective.

It changes how investors should value these companies, too. Defense businesses used to be valued on how unique the platform was. Increasingly, they’ll be valued on how quickly they can improve it. The winners won’t automatically be the ones with the most advanced hardware; they’ll be the ones with the shortest distance between a problem discovered in the field and a solution deployed back to operators. The fastest learning loop becomes the competitive advantage, and the companies that design for that from day one will have an advantage that’s difficult to replicate.

The industry is finally internalizing a discipline software companies have followed for decades: ship, learn, improve, repeat. That doesn’t turn defense companies into SaaS businesses. It means software is becoming inseparable from the value of every defense platform.

For investors, founders, and defense leaders, that’s where competitive advantage is moving. Deployment used to mark the end of development. In the next generation of defense technology, it marks the beginning. Every platform deployed is simply Version 1.0.