AI, Government, and the Future

The episode delivers substantial strategic and technical insights about AI deployment in maritime defense, including novel framing around the 'third offset' technology, the hedge strategy concept, and practical approaches to AI integration via condition-based maintenance and tactical PaaS architecture. However, it mixes significant content with some meandering storytelling (the Dennis Connor anecdote, extended Naval Academy background) that dilutes density. Most claims are substantive but some lack depth.

“The largest risk for large scale conflict in the world is failed deterrence on the part of the United States because our adversaries, they can win conventional war.”

“what we've built isn't really just a shipping. What we've built is a manned node inside a learning ecosystem of maritime power that surrounds, subsumes and overwhelms your ships with electronic fire control.”

Staples offers genuinely fresh frameworks including the hedge strategy analogy (high-end platforms + small agile systems), the 'better acuity vs. autonomy' framing, and the idea of using proven OT&E/DT&E processes to govern AI safety testing. The trolley problem applied to friendly fire and the Sears-vs.-Amazon distinction are well-executed but not entirely novel. Some arguments (Internet/DoD underoptimization, nuclear submarine lessons) are recycled historical examples, though effectively deployed.

“Sears with a website is not Amazon. Right. So there is a way that you're like, hey, we're going to go get an AI algorithm on a ship and what you end up with is sears with a website.”

“what exactly is it going to take to get the other hundred out? So what we've done, again at the highest strategic level is we told the Chinese, well, you can read in James how many ships and submarines we have, and you've really only got to get ready to beat them out of third.”

Zac Staples is a highly credible operator with 23 years of active Navy service, direct experience integrating AI on Navy warships, and current CEO of a defense AI firm. He speaks from deep operational knowledge rather than theory. His references to collaboration with Admiral Daryl Cotton, Laura Selby, and unnamed ethicists at the Naval Postgraduate School further establish credibility as a connected practitioner at the strategic level.

“I had the distinct honor of serving our country in the Navy for, uh, almost 23 years.”

“I had several things, you know, innovative things that I really enjoyed. And you know, while I was on active duty and, and then I'm like, I'm building the company to get after this, how to actually build and deploy this set of third offset tech and make it work with the existing systems.”

The episode provides some concrete examples (300 ships, 100 deployed / 100 in maintenance / 100 intermediate; 6-7% AI error vs. 10% friendly fire baseline; 80% of iPhone tech from DARPA) but often retreats into abstraction. The condition-based maintenance program and tactical PaaS discussion lack named examples or metrics. Strategic claims about China's manufacturing and U.S. aircraft superiority are asserted without numbers. More data points and actual case studies would strengthen the argument significantly.

“Admiral Darryl Cottle is the commander of the U.S. he's the commander, of course, and a couple of weeks ago he was explaining it like he was like, I've got about 300 ships and suckers right. When you, when you add them all together, he goes, at any given time, 100 of them, um, are overseas, are out at conducting four presence operations”

“if we know that there is some percentage of uh, friendly fire incidents and let's just say it's 10% in a particular type of casualty and the AI in every simulation we could possibly put it through is about 6 or 7%”

Max Romanick asks solid strategic questions and follows up thoughtfully on test-and-evaluation frameworks, but rarely pushes back or probe soft claims. When Staples makes big assertions (e.g., China can beat 1/3 of the Navy), the host accepts them without challenging sourcing. The interview is conversational and warm but lacks adversarial rigor. Good thematic threading (legacy system risks, interim strategy) but limited willingness to call out vagueness or demand specifics.

“What I really like is that you're setting up this sort of like one of my questions for you is like, okay, so how do we mitigate this risk?”

“I mean it's the trolley problem on sterilites a hundred percent.”