All podcasts / All-In Podcast / Summary

China Outbuilds America 230-to-1. Saronic Has a Plan

2026-08-06 - 48 min - source - Read full transcript
Jason Calacanis (host)Dino MavrogenisVibhav Altekar

Key insights

China can build ships 230 times faster than the US by tonnage.
The US produces about 100,000 gross tons of shipping capacity per year; China produces roughly 23 million gross tons, a 230-to-1 gap. Thirty years ago China held only 5% of world shipbuilding capacity; today it holds 57%, and that share is still rising.
shipbuilding-gap
Beijing's shipbuilding dominance was built by subsidizing the entire commercial industry, not just the military.
China subsidizes direct build costs plus raw materials and labor, undercutting commercial shipbuilders on price globally. As a result a ship built in the US costs five to six times what the same ship costs to build in China, so China keeps winning more of the world's commercial orders.
shipbuilding-gap
The US Navy fleet is shrinking even as the threat grows.
The Navy has 296 ships against a congressionally mandated minimum of 355 (set in 2018). Last year the US built 9 ships but retired 19. China's navy has grown from a 'littoral backwater' 30 years ago to roughly 370 ships including nuclear submarines, carriers, and hypersonic missiles, expected to reach 450 soon.
shipbuilding-gap
The commercial shipbuilding gap is even more extreme than the military one.
China delivered over 1,000 commercial ships last year; the US delivered five. In a real conflict, commercial shipbuilding capacity is what converts to wartime production (as it did for the US in WWII), so this commercial gap is treated as the deeper strategic vulnerability, not just naval fleet counts.
shipbuilding-gap
Removing humans from ship design collapses cost and complexity, not just weight.
Autonomous vessels eliminate berths, commissaries, latrines, separate human-rated electrical systems, doors, stairs, and shock-mitigation requirements. Saronic's Corsair is a 24-foot autonomous boat that can travel roughly 1,000 nautical miles and outperforms a comparable manned commercial 24-foot boat because it isn't designed around a human occupant.
autonomous-warfare
Autonomous fleets change naval economics by orders of magnitude, measured in VLS tubes fielded per year.
A single manned destroyer costs about $3 billion and 6-8 years to build, fielding roughly 96 VLS (vertical launch system) missile tubes at about 10-15 tubes/year of fielded capacity. Saronic's 180-foot autonomous Marauder can carry the equivalent of 16 VLS tubes each; building 20 Marauders/year fields 320 VLS tubes/year at a fraction of the cost.
autonomous-warfare
Cost-plus contracting structurally rewards legacy primes for running over budget and behind schedule.
Under cost-plus, a contractor earns roughly 10-15% margin on whatever the program ends up costing, so a $100M program nets $15M but a $1B program nets $150M - creating a direct incentive to let costs balloon. Saronic instead self-funds R&D with $2.5B in private capital raised over about four years and sells to the government on firm-fixed-price contracts.
defense-procurement-reform
Shipbuilding has historically separated design from manufacturing, entrenching single-source suppliers and slow iteration.
Roughly 70% of sequence-critical components on some Navy ships come from sole-source suppliers, and the firm that designs a ship is typically different from the one that builds it, removing any incentive to redesign for speed or cost. Saronic collapses this by owning both the design and the shipyard, letting software engineers, welders, and naval architects work together directly.
defense-procurement-reform
Only about 1% of the Department of War's budget currently goes to autonomous systems, which the founders argue is far too low.
Dino states this figure directly and argues the share needs to rise quickly (without specifying an exact target, floating '5%' as an example) given the scale of the shipbuilding and fleet-size gap with China.
defense-procurement-reform
Saronic is building 'Port Alpha,' a new mega-shipyard in Brownsville, Texas, designed from scratch to co-locate ship design and manufacturing.
The company already operates a 100-acre shipyard in Franklin, Louisiana producing Marauders (ramping toward 50/year). Port Alpha starts on an 800-acre site in Brownsville (with room to scale to 4,000 acres), backed by billions of dollars of investment and expected to create 10,000 jobs over 10 years, chosen partly for its proximity to SpaceX's Starbase and Houston/San Antonio's industrial and software labor pool.
us-industrial-revival
A Saronic Corsair carried out the first known autonomous-platform rescue of downed military pilots, in the Strait of Hormuz.
Two US pilots went down in a contested environment during a conflict with Iran; Saronic's Corsair, an unmanned 24-foot autonomous vessel, was sent to retrieve them without putting additional personnel at risk. Dino, an 11-year Navy SEAL veteran (including 5 years on SEAL Team 6), frames it as proof the US no longer has to risk personnel to avoid leaving people behind, drawing a contrast to a 2005 mission where a rescue helicopter was shot down retrieving trapped SEALs in Afghanistan.
autonomous-warfare
Autonomous weapons still route life-and-death decisions through human-set policy, not machine discretion.
Vib describes a US standard ('3009') that governs how AI can be used on autonomous systems: the AI flags a target and recommends action, but a human sets the engagement policy and approval thresholds in advance (e.g., in a full Taiwan Strait conflict, 'anything that's not a friendly vessel is an enemy vessel'). This is framed as fundamentally different from China's approach, which the guests say isn't bound by comparable treaties or ethical constraints.
autonomous-warfare

Companies

Techniques and frameworks

Summary

Jason Calacanis interviews Saronic co-founders Dino Mavrogenis (CEO, and an 11-year Navy SEAL veteran including five years on SEAL Team 6) and Vibhav Altekar for the All-In Interview series. The episode opens with a striking real-world story: Saronic's Corsair, a 24-foot fully autonomous vessel, was used to rescue two downed American pilots in the Strait of Hormuz during a conflict with Iran - reportedly the first time an autonomous platform has carried out a manned-pilot rescue. Dino frames it against a personal memory of a 2005 mission in Afghanistan where a helicopter was shot down attempting to rescue trapped SEALs, arguing the moment proves the US can honor its "leave no one behind" ethic without risking additional personnel.

The core of the conversation is the shipbuilding gap between the US and China. The US builds about 100,000 gross tons of ships per year; China builds roughly 23 million, a 230-to-1 gap. China's share of world shipbuilding capacity rose from 5% thirty years ago to 57% today, driven by state subsidies that undercut commercial shipbuilders globally and make a US-built ship cost five to six times more than the same ship built in China. The US Navy fleet has shrunk to 296 ships against a statutory minimum of 355, and last year the US retired more ships (19) than it built (9). The commercial gap is even starker - China delivered over 1,000 commercial ships last year to the US's five - which matters strategically because commercial shipbuilding capacity is what converts to wartime production in an actual conflict.

Saronic's answer is autonomy plus vertical integration. Removing the human from a vessel eliminates entire categories of cost and complexity - no berths, life support, doors, stairs, or shock-mitigation systems - letting a 24-foot autonomous Corsair outperform a comparably sized manned commercial boat. The founders reframe naval procurement around VLS (vertical launch system) missile-tube economics: a $3 billion manned destroyer takes 6-8 years to field about 96 tubes, while a fleet of 20 autonomous Marauders per year can field 320 tubes annually at a fraction of the cost. They contrast this with the cost-plus contracting model that has historically paid legacy defense primes a straight percentage margin on top of program cost - an incentive structure that rewards delay and overrun. Saronic instead self-funds R&D (having raised $2.5 billion in private capital over about four years) and sells to the government on firm-fixed-price contracts, alongside peers like Anduril.

The episode includes a news announcement: Saronic is building "Port Alpha," a new shipyard in Brownsville, Texas, starting at 800 acres (with room to scale to 4,000) and billions of dollars of investment, expected to create 10,000 jobs over ten years. It's designed from the ground up to co-locate ship design and manufacturing - unlike the traditional industry, where design and build are handled by separate companies with limited incentive to optimize either for the other - drawing an analogy to how chip companies co-design silicon and compilers. Brownsville was chosen partly for its proximity to SpaceX's Starbase and to Houston/San Antonio's blended software and industrial labor pool.

On the AI-and-autonomous-weapons question, Vib describes a governing standard ("3009") under which AI recommends targeting but engagement authority and thresholds are set by human government policy in advance, contrasting this with China's approach, which the guests characterize as unconstrained by comparable treaties. The conversation closes on the founders' broader thesis: Saronic sits within a wave of venture-backed defense-tech companies (alongside SpaceX, Anduril, and Palantir) that they see as filling a strategic gap the government previously depended on legacy primes to cover, and Dino describes the moment as "our generation's version of the space race."

Notable Quotes

"The United States can build 100,000 gross tons of ships every year... The Chinese can build 23 million gross tons. So they can outbuild the U.S. 230 to one." - Dino Mavrogenis

"Last year, we built nine ships, but we retired 19 naval vessels. So we're going in the wrong, the completely wrong direction." - Dino Mavrogenis

"I'm actually incentivized, and I'm not saying there's malicious intent, but I'm incentivized to make it take longer and cost more, because if it costs $100 million, well, then I only make $15 million. But if it costs a billion dollars, then as a company, I make $150 million." - Dino Mavrogenis, describing cost-plus contracting

"I tell our team, I'm like, this is our generation's version of the space race, make no mistake about it, and we cannot lose." - Dino Mavrogenis