2025-05-06

Aliens, AI Weapons, China & Global Conflict: Palmer Luckey Sounds the Alarm | EP #169

Palmer Luckey joins Peter Diamandis for his fourth Moonshots appearance, opening with UFO/NHI testimony he can't fully discuss before pivoting to how Anduril is building toward digital superintelligence through its Lattice AI platform. He argues the real AI-era danger is less 'US vs China' and more individual rogue actors, while China's revisionist territorial claims and roughly 300x shipbuilding advantage make Taiwan the central flashpoint. Palmer defends case-by-case autonomous-weapons ethics, previews Anduril's $900M Columbus fighter-jet factory and newly public Pulsar L drone-killing system, and predicts subterranean warfare as the next military domain. The conversation closes with a rapid-fire AMA covering robotics form factors, Bitcoin, Mark Zuckerberg, and Palmer's philosophy on wealth, eccentricity, and raising kids in the age of Fortnite.

▶ Watch on YouTube

Topics

Palmer Luckey on UFO/NHI congressional hearings Other ▶ 0:16

Palmer says he's seen unexplained things he can't discuss but no proof of recovered alien craft; he floats that the handful of truly inexplicable cases are more likely time travel or an unknown natural phenomenon than nearby alien visitors.

Palmer's alien-consciousness thought experiment Other ▶ 2:58

Riffing on human perceptual lag, Palmer describes a sci-fi idea he once wrote: a being whose consciousness runs far slower than ours but reasons so far ahead that its responses still look instantaneous, making it undetectable until something goes wrong.

Anduril's AI-first strategy and the road to digital superintelligence AI ▶ 14:19

Palmer explains Anduril was built on the early bet that AI and autonomy would become real; its Lattice platform's goal, inspired by Laplace's Demon, is near-omniscient prediction of the battlefield, which he values more than faster hardware.

US and China vs the rogue actor Geopolitics ▶ 35:07

Pushing back on Peter's 'US and China vs rogue actors' framing, Palmer argues nation-states act rationally and are less terrifying than individual rogue actors who might see their own death or an apocalypse as victory, like a tailored bioweapon.

China's revisionist territorial threat Geopolitics ▶ 36:58

Palmer argues China is a unique threat because it rewrites history (claiming Okinawa, parts of the Philippines, Vietnam) to justify eventual force, and it has made explicit threats to retake Taiwan by force within this generation.

Anduril going public and the founder-led company advantage Economy ▶ 21:06

Palmer confirms Anduril will eventually go public, modeling Tesla's strategy of attracting investors who believe in the long-term vision and deliberately repelling short-term-focused ones, so going public doesn't force the company toward safe, low-R&D decisions.

What killed Cold War-era defense manufacturing speed Economy ▶ 43:51

Comparing WWII-era build speeds (a new jet in 143 days, Liberty ships cut to 4.5 days) to today's slow defense programs, Palmer blames the post-Cold-War 'Last Supper' consolidation that reduced competition and let inefficiency become politically entrenched for decades.

Anduril's design philosophy: no vertical integration, mass-producibility Other ▶ 48:36

Unlike SpaceX or Tesla, Anduril deliberately avoids vertical integration so it can multisource parts and 10-100x production overnight if the DoD suddenly needs it, borrowing fast-iteration habits from consumer electronics rather than legacy defense contracting.

Case-by-case ethics of autonomous weapons AI ▶ 1:14:17

Palmer argues blanket bans on AI-enabled weapons are a mistake; the real question is minimizing civilian casualties (e.g. a landmine smart enough to not kill a school bus), with human accountability for every lethal decision as the non-negotiable line.

Wildfire detection and suppression technology and the XPRIZE Other ▶ 1:23:40

Palmer and Peter discuss Anduril's early autonomous firefighting work (killed by political pushback over job losses) and argue the tech to detect and stop roughly 95% of destructive wildfires already exists; it's now a cost and deployment problem, central to the $11M XPRIZE Wildfire.

Subterranean warfare as the next military domain Other ▶ 1:05:38

Palmer predicts the crust of the earth will become a full 3D battle space for logistics and combat, following Cold War-era US/Soviet earth-boring vehicle programs, eventually warranting its own dedicated military branch like the Space Force.

Pulsar L drone-killing electronic warfare system Robotics ▶ 1:08:21

Anduril publicly showed Pulsar L, a cooler-sized AI-powered EW system that downed 25 live autonomous attack drones in a real test; it's already deployed in combat, but is illegal for civilian use in the US and Palmer expects that to change.

Predictions made

open Palmer Luckey: Anduril's autonomous fighter jets, now in construction at a $900 million Columbus, Ohio factory, will be flying in actual combat.

· due: 2029 · ▶ watch

“Those are going to be in combat before 2030.”

Your call:

open Palmer Luckey: Anduril will eventually go public.

· due: unspecified · ▶ watch

“we are we are going public... can't win an F-35 scale program without doing”

Your call:

open Elon Musk: AI will become as smart as all of humanity combined.

· due: 2029-2030 · ▶ watch

“as smart as all humans combined by 2029 or 2030”

Your call:

open Palmer Luckey: Warfare will look largely similar to today, dominated by legacy Cold War-era platforms, with only a small pocket of breakthrough AI capability like a thin tip-of-the-spear of autonomous jets.

· due: 2030 · ▶ watch

“warfare in 2030 is going to look more or less the same as it does today”

Your call:

open Palmer Luckey: China will be ready to fight the US militarily, making readiness by this date the internal mandate ('China 27') driving which near-term weapons Anduril works on.

· due: 2027 · ▶ watch

“if the feature you're building... is not going to be ready for a fight with China before the end of 2027, you can't be working on it”

Your call:

open Palmer Luckey: Subterranean warfare will become a major military domain, eventually requiring its own dedicated branch similar to the Space Force.

· due: unspecified · ▶ watch

“I think it'll become a large enough part of warfare that you're going to need a dedicated group”

Your call:

open Palmer Luckey: US regulation will loosen to allow counter-drone systems like Pulsar L to be used domestically (airports, stadiums), not just by the military.

· due: unspecified · ▶ watch

“I think that's going to change”

Your call:

open Palmer Luckey: A large-scale terror attack or series of attacks using drones will eventually happen in the US before defenses are widely deployed.

· due: unspecified · ▶ watch

“I suspect inevitably there will be someone who commits a large scale terror attack or series of terror attacks using drones”

Your call:

open Palmer Luckey: Technology already exists to detect and stop roughly 95% of destructive wildfires; it just needs to be assembled and deployed at scale.

· due: unspecified · ▶ watch

“The tech to detect and exterminate destructive wildfires... it exists today”

Your call:

Numbers that matter

GPT-o3 scored a 135 IQ; Claude 3 was the first AI to score just above average human IQ at 101 Cited by Peter while framing the pace of AI intelligence gains.
Anduril's job applications rose 3x in the week after its deliberately off-putting 'Don't work at Anduril' recruiting campaign Palmer's example of attracting the right people by repelling the wrong ones.
WWII-era Liberty ship production time was cut from 230 days to 4.5 days per ship Cited alongside Kelly Johnson's 143-day first US jet and the P-51's 102-day concept-to-flight timeline as examples of wartime manufacturing speed.
B-24 Liberator bombers were built at a rate of one every 63 minutes during WWII Peter's example of how fast US industry once iterated on weapons production.
Anduril is building a $900 million factory in Columbus, Ohio to manufacture autonomous fighter jets Palmer says the jets built there will be in combat before 2030.
Anduril's Pulsar L system downed all 25 autonomous attack drones in a live test by jamming/spoofing them out of the sky Palmer describes the test as real, not CGI, after public skepticism about a released demo video.
XPRIZE Wildfire carries an $11 million prize with evaluations scheduled for October 2025 Anduril was the first registrant for the prize.
China has about 300 times more shipbuilding capacity than the United States Cited as a key amphibious-invasion advantage for a potential Taiwan campaign, independent of wartime surge capacity.
XPRIZE Carbon Removal awarded $100 million across 1,300 teams from 88 countries; winner Mati Carbon received $50 million Mati Carbon's rock-weathering (basalt dust) technique also boosts crop yields 20-30% by improving water retention.
Anduril now has nearly 5,000 employees Palmer cites this scale while explaining why executives can't simply override multi-level personnel decisions.

Worth digging into

🕳️ Jim Bell's 1996 'Assassination Politics' essay

Palmer cites it as predicting Bitcoin and crypto's societal impact decades early, written by a fringe figure later imprisoned as a terrorist and tax evader.

🕳️ Soviet subterranean warfare vehicle program

Palmer claims the USSR built and then lost track of a working earth-boring vehicle during the Cold War, and argues the crust of the earth will become a full 3D battle domain.

🕳️ Anduril's Pulsar L public demo and its real combat use

Palmer says the viral 25-drone takedown video was a real test, not CGI, and that the system is already deployed with military customers, but domestic civilian use is still illegal.

🕳️ The 'Last Supper' defense industry consolidation

Palmer blames a 1990s Pentagon-driven consolidation of defense contractors for killing the WWII-era pace of weapons innovation for decades, a direct historical precedent for Anduril's own thesis.

🕳️ Mati Carbon's rock-weathering scale-up

The $50M XPRIZE-winning technique spreads basalt dust on farmland, capturing carbon while boosting crop yields 20-30%, and Peter says a philanthropist is about to fund a 100x scale-up.