The $65 million cube

Celebrating SPDHTs

In today’s Diary:

• SPDHTs

• Mining asteroids with lasers

• 100,000 war drones from scratch

• Robots that make it rain

• Around the world in 7 hours

Dear Rational Optimist,

Imagine a cube 14 inches on each side, about the size of a microwave.

At 2,200 pounds, no man can lift it. And it’s worth $65 million.

AstroForge is on a mission to bring these back from space and sell them.

You know how the last 40 to 50 years of innovation have been primarily digital?

We got smartphones, social media, and streaming. But as Peter Thiel says, “Where are the flying cars?” Why hasn’t our physical world improved much—and in some ways has even deteriorated?

One reason is our smartest people were flocking to companies like Facebook and Google, using their ample brainpower not to build real things, but to increase ad revenue.

Fresh off our Rational Optimists Take America tour, I am happy to confirm there’s been a shift in the zeitgeist. We met over 30 Smart People Doing Hard Things, or SPDHTs for short, who are tackling real, physical problems.

Want to meet 5 SPDHTs while they’re still young and under the radar?

AstroForge (north of LA)

Smart person: Matt Gialich

Hard thing: Mining asteroids

Why it’s important: We have plenty of platinum group metals (platinum, palladium, etc.) here on Earth. Problem is they’re deep underground, making them dangerous and extremely expensive to extract.

It turns out many asteroids orbiting Earth are rich in these valuable metals. AstroForge Founder Matt Gialich told us asteroids contain ore that’s 5,000 to 10,000 times purer than the best mines in South Africa.

Gialich’s thesis: Advances in rocket technology—plus SpaceX’s creation of the blossoming space economy—have made it economical to extract valuable metals from these asteroids and fly them back to Earth.

Here are Stephen, Gialich and I under a half-scale replica of AstroForge’s first exploratory space vehicle:

Because platinum metals are so valuable, you don’t need to harvest much to make a lucrative business. One 14-by-14-inch cube is worth $65 million!

AstroForge is the perfect example of Matt Ridley’s “ideas having sex.” Rocket scientists work side-by-side with geologists.

AstroForge’s space vehicles hitch a ride on a rocket. Once in space they detach, fire up their thrusters, and fly to a preselected asteroid past the moon and about 10% of the way to Mars. The vehicle matches the asteroid’s speed and lands on it.

Then comes the hard part.

Matt explained to us the metal must be refined to a high purity on the asteroid. Otherwise they’d have to bring back tons of raw material, which ruins the economics.

They first tried using magnets to refine. But the magnets needed more power than the on-board solar panels could deliver.

Now they’re using lasers to refine… and it’s working.

AstroForge is one of many, many businesses made possible by SpaceX’s creation of the commercial launch market.

This is an audacious company with an outrageous goal. Rational Optimists are cheering you on, Matt.

Neros (El Segundo)

Smart person: Olaf Hichwa

Hard thing: Mass-producing cheap drones in America

Why it’s important: America is alarmingly behind China in drones. If drones decide the next war, we lose.

Drones have already redefined war in Ukraine. For reasons we Deep Dove into here, drones will only get more asymmetrically important as autonomous swarms arrive.

Picture swarms of thousands of AI-powered drones overwhelming an aircraft carrier or base. We’re about 5 years away from this, maybe less.

China made about 8 million drones last year. The biggest US drone maker, Skydio, made only 10,000.

Can’t we just make more?

Catching up is hard because most drone parts—batteries, sensors, cameras, motors—are made in China’s innovation capital of Shenzhen. Neros co-founder Olaf tells us: You can walk down the street in Shenzhen and buy a drone part that’s better than any US-made part for 1/10^th the price.

Neros is making US factories to mass-produce cheap, lightweight war drones. It has already eliminated Chinese parts from its drones and can pump out 100 drones a day. Its goal is to make 100,000 drones next year. That will make Neros America’s largest drone producer by an order of magnitude.

Neros just raised $75 million from top VC firm Sequoia to help make it happen. In the warehouse, Olaf showed us $20 million of inventory that sits waiting to be assembled:

Excuse Stephen’s CrossFit outfit, we went straight from the airport on Saturday.

Neros is doing the very difficult work of developing mass manufacturing of drones from scratch. It’s hard to think of something more important.

Rainmaker (El Segundo)

Smart person: Augustus Doricko

Hard thing: Rain on demand

Why it’s important: Many places that need water don’t have it.

Did you know Utah’s Great Salt Lake is drying up? It has only a quarter of the water it had 100 years ago, and may not exist in five years without intervention.

Augustus Doricko’s Rainmaker is intervening. Rainmaker has begun the largest cloud-seeding project in American history to refill the Great Salt Lake. Specifically, it’s restoring the snowpack in the Bear River Basin upstream of the Great Salt Lake. Rainmaker’s initial goal is to bring 10 billion gallons of new water to the lake.

Cloud seeding is a proven technology invented in the US in the 1940s. It works by spraying a substance, usually silver iodide, into a cloud under the right conditions. Water in the cloud then freezes to tiny particles and falls, creating rain or snow.

Rainmaker is using drones to cloud seed, which allows it to be far more precise than ground-based systems or manned aircraft.

Augustus shows us a video of how one speck of ice on a drone propeller causes it to snap off. That’s no good when deliberately flying drones into storms like Rainmaker does.

So Augustus and his team invented a new way to de-ice drone propellers. He showed us the de-icing chamber where it’s tested. He also showed us the robotic “den” they created for drones to land, recharge and refuel during missions. After initial setup, Rainmaker’s system can all but run itself.

Rainmaker’s biggest problem isn’t tech, it’s public relations. Many think controlling the weather is “playing God.” Others think the silver iodide that’s sprayed into clouds is dangerous. Augustus spends as much time dealing with bureaucracy as on drone tech.

Cloud seeding is banned in Florida and Tennessee. No coincidence that these are two of the six rainiest states in the country. Utah, the second driest state, needs innovation.

Augustus is the SPDHT face of Gundo. He’s unabashedly pro-American, a Berkeley dropout and Thiel Fellow. We visited Rainmaker’s offices on a Sunday. But you wouldn’t know it was a weekend. The office was packed with motivated employees tinkering with Rainmaker’s in-house radar system.

Aalo Atomics (Austin)

Smart person: Matt Lozak

Hard thing: Mass-producing Small Nuclear Reactors (SMR) efficiently

Why it’s important: We need a lot of new energy (preferably clean) to power AI—and we need it yesterday.

This was our third visit to meet Aalo Atomics Founder Matt Lozak. The rapid growth is remarkable.

One year ago the warehouse was empty. This time it was buzzing with welders and trucks and a fully assembled test reactor. Here we are in front of it with Matt Ridley:

Aalo is making a sodium-cooled reactor. Traditional reactors are water-cooled, which introduces risk as water turns to steam and builds explosive pressure.

Aalo’s unique positioning is “making the factory the product” for SMRs. By standardizing production—building reactors assembly-line style, then shipping them to sites—Aalo aims to drive down costs and startup time.

Its long-term goal is to build 100 reactors a year. That’s a reactor every 3 days—quite an improvement from the 10+ years it takes to build an old-school nuclear plant in the US!

We lampoon nuclear regulators a lot around here, and they deserve it. But Matt told us the biggest tailwind in nuclear isn’t regulatory change. It’s AI.

Traditionally, nuclear had only one customer—utilities, which are risk-averse and slow to embrace new technology. Now, companies that operate AI data centers are lining up and willing to pay a premium for the clean, always-on power that SMRs provide. This new source of demand has changed everything.

I remain convinced Aalo will turn on its first reactor by July 4, 2026, as Stephen and I predicted here.

Astro Mechanica (San Francisco)

Smart person: Ian Brooke

Hard thing: Building a new supersonic jet engine

Why it’s important: So we can fly anywhere on Earth in a few hours, as we explored in our Deep Dive here.

Founder Ian Brooke is the ultimate SPDHT. Rather than retooling existing technology for supersonic flight, he’s building a whole new engine from scratch.

While traditional jet engines are one-speed-fits-all, Ian’s new turboelectric adaptive engine is a chameleon.

For takeoff and landing it’s an efficient turbofan, sipping fuel. When it goes supersonic it morphs into a powerful ramjet. To potentially go hypersonic (5 times the speed of sound) in the future, it will transition into a rocket-like mode with no moving parts.

Here’s Matt Ridley with Ian in front of the engine:

Ian’s goal for Astro’s first jet: fly five passengers anywhere in the world… in under seven hours… for cheaper than your average commercial flight today.

How is that even remotely possible? Another breakthrough: Astro’s engine will run on liquefied natural gas (LNG), which is a quarter the cost of traditional jet fuel and 60% denser.

I’ll stop there because Stephen is working on a deeper dive into Astro Mechanica versus Boom Supersonic in the race for supersonic supremacy. More soon.

Don’t forget your #1 job as an ROS member: Help us free more Rational Optimist minds by clicking like and restack below and forwarding this to your friends!

Follow me on X (Twitter) here.

—Dan Steinhart

---

Dan Steinhart is a co-founder of the Rational Optimist Society.

---