Good evening, and welcome to Two to the Fifth, the ten-minute news show for people who think news should probably clear a slightly higher bar than “this happened eleven minutes ago.” Most things that happen today won’t matter very much. Some become more important with time.
So we look at five timeframes: the last 24 hours, the last week, the last month, the last year and the last decade. Two minutes each. Five pours. One country. No chaser.
The goal is pretty simple. Ten minutes here should leave you with a clearer picture of the world than you had after scrolling two hours of headlines. This week: submarines, nuclear plants, missiles, pond scum, factory jobs, HIV and houses. We think there’s something in each worth paying attention to.
Decide for yourself, but first gimme 10 minutes.
This is Two to the Fifth.
THE LAST 24 HOURS
Sure, I Have $4 Billion for That
Two enormous investments landed almost on top of each other.
Defense-tech company Anduril announced a $3.7 billion manufacturing complex at Sparrows Point, Maryland, the old Bethlehem Steel site. The plan is to make parts for the Navy’s submarine industrial base and create more than 3,000 permanent jobs.
About the same time, Google signed a huge power agreement with Constellation Energy. Google wants 3.6 gigawatts of electricity for data centers. Constellation says that demand will support more than $4 billion in investment, including upgrades at 11 nuclear reactors.
Submarines. Google searches. Not a lot in common. Except somebody with a very large checkbook just said, we’re going to need a whole lot of this for a long time. And suddenly billions of dollars worth of capacity starts looking like a reasonable investment.
We spend a lot of time asking why companies don’t build things here anymore. Apparently one useful answer is, give them a reliable customer.
What If Vaccines Didn’t Need a Refrigerator?
For all the worrying about AI taking our jobs, or ending civilization as we know it, here’s something it’s actually doing today. Reuters reports that researchers used AI to redesign the tiny lipid particles that carry mRNA vaccines. The work, published in Nature Biotechnology, produced experimental formulations that stayed effective for up to a year at room temperature, and two months at nearly 100 degrees. mRNA vaccines normally need cold storage and transport, which makes them expensive and difficult to get to places that don’t have reliable refrigeration.
Still experimental. So far tested in animals. But if it works in people, AI didn’t replace anybody’s job - it helped someone do their job better. And it may have helped vaccines reach millions more people.
THE LAST WEEK
Can You Make More Missiles, Like, Now?
Last week the Navy gave Raytheon a contract worth up to $24.4 billion to make more Standard Missile-6s. That follows a $20.7 billion agreement for AMRAAM missiles and a $58.6 billion deal for Patriot interceptors.
Why all the urgency? Because Ukraine and the wars in the Middle East, particularly Iran, have been chewing through some sophisticated weapons faster than America can replace them.
The United States spends hundreds of billions of dollars every year on defense. We protect domestic weapons production. We pay extremely well for some of this stuff. And we still discovered that when an actual war starts using missiles at scale, making enough new ones can take years.
Apparently having the money to buy something and having the ability to make enough of it are different problems.
Pond Scum Wins the Nobel Prize
Meanwhile, this year’s Nobel Prize in medicine went to three scientists for helping develop optogenetics. Which sounds like something involving self-replicating bifocals. Don’t worry, Warby Parker. It’s not.
In simple terms, scientists figured out how to use light-sensitive proteins to turn specific brain cells on and off. And the story starts with researchers studying how single-cell algae in your local pond know which way the light is coming from. Those nifty little proteins eventually became tools that let neuroscientists activate particular brain circuits with flashes of light.
Brain circuits involved in movement. Memory. Parkinson’s. Depression. Vision. All kinds of things we’re still trying to understand.
So the next time somebody asks why we spend money on stuff with no obvious commercial application, show them your Nobel. And don’t wipe off the algae first.
THE LAST MONTH
The Factories Are Coming Back, Just Not All the Jobs
There really has been a manufacturing investment boom in America. Chip plants. Battery factories. Energy equipment. Medical products. But here’s what didn’t come back at anything close to the same rate. The jobs.
American manufacturing employment today is around 12½ million people. Back in 1979 it was almost 20 million. Turns out a semiconductor plant in 2026 doesn’t employ people the way a steel mill did in 1976. Why? Automation. Robotics. Software.
None of that is bad. But it does mean “manufacturing is coming back” needs a follow-up question. Which part? The investment? The factory? The output? Or the jobs? Because those are not necessarily the same thing.
We Finally Beat the Dust Bowl. Yay.
For almost 90 years, the summer of 1936 was America’s shorthand for brutally hot weather.
The Dust Bowl. Crop failures. People packing everything they owned into whatever still ran and heading west. One of those historic summers your grandparents could use to shut down any conversation about how hot you thought it was.
Well, congratulations everybody. We beat it. This summer became the hottest on record across the Lower 48. July was the hottest single month since we started keeping continuous track of it in 1895.
I get it, records were meant to be broken. I’m just not sure anyone was rooting for this one.
THE LAST YEAR
This Is What a Manufacturing Comeback Looks Like
A few years ago, some of the biggest new manufacturing bets were electric vehicles and batteries. Then demand expectations changed. Federal policy changed. Tariffs changed. Projects got delayed or canceled. At the same time, money started pouring toward other things. AI infrastructure. Making electricity. Semiconductors. Medical supplies.
This is what a comeback looks like. Fits and starts. Winners and losers. Companies make manufacturing bets when they believe somebody is going to buy what comes out of the factory five or ten years from now. Opening a factory is a headline. Operating it at capacity over the long run is the more important part.
Teaching the Body to Control HIV by Itself
One of the quieter medical stories of the past year came from a small HIV study. Researchers took patients’ own immune cells, genetically modified them to recognize HIV and resist infection, then put them back.
Ten people received the treatment and then stopped their regular HIV medication under medical supervision. Most saw the virus rebound. But one showed no detectable rebound during the planned 16-week interruption, and another kept it at bay for about 90 weeks without medication, with several others showing signs of partial control after the virus returned.
Three people is not a cure. You won’t see it at CVS next Tuesday. But for decades the great breakthrough in HIV was “take a whole bunch of medicine every day for the rest of your life, and you can turn a fatal disease into something people can live with.” Now research has moved to “can we teach the immune system once and let it keep doing the job?”
That’s a very small study tackling a very big question.
THE LAST DECADE
What About Those Factory Jobs?
Since manufacturing employment peaked in 1979, America has lost roughly seven million factory jobs. We’ve talked about those lost jobs for decades. But most Americans still have jobs. In fact, America created tens of millions of them in healthcare, professional services, construction, logistics, finance, technology, restaurants and management.
Jobs didn’t so much disappear as change. Some people did great in that transition. Others didn’t. If Dad lost a well-paid union factory job and wound up working for much less in a warehouse, the new economy did him no favors. If his daughter became a nurse or software developer and earned considerably more, that same transition looks different. Some cities found that next economy, while others spent thirty years driving past the factory where the last one used to be.
Which raises a different question than “How do we bring the old manufacturing jobs back?” It’s “what did the good ones actually do for us?” They let somebody without a college degree become skilled, useful, well-paid and middle class. We don’t necessarily need the 1979 factory back. But we could sure use that function.
So What Could We Manufacture That We Don’t Now?
The Industrial Revolution was basically about taking things made one at a time and figuring out how to make lots of them in one place, better and cheaper, then ship them to the customer.
Henry Ford famously did that with cars, helping turn them from expensive toys into a necessity ordinary families could own.
So what’s a necessity today that too many ordinary families can’t afford? Housing. And interestingly, it’s still one of the few gigantic things we largely build one at a time, outdoors, exactly where the customer is going to use it.
Imagine Ford sending thirty people to your driveway with steel, wiring, glass and rubber and saying, “We’ll have your Explorer ready in six months. And it’s going to cost a lot more.” Yep, ridiculous. Also kind of how we build houses.
To be fair, we already manufacture parts of them. Roof trusses arrive preassembled. Same with windows, doors, cabinets and increasingly floor and wall systems.
And we’ve done it at scale before. Levittown was basically an assembly line spread across a subdivision. A few designs built by specialized crews using repeated processes. Once the owners moved in, they could spend the next fifty years making them unique.
So the question isn’t whether houses can be manufactured. It’s why, for the most part, they aren’t. Part of the answer is standardization. Housing has different codes and approvals depending on where you live. Part of it is predictable demand. Housing has lots of individual, unpredictable customers.
So imagine a city or state saying, “We need 20,000 starter homes. Here are several approved designs. Here are the rules. Here’s enough demand to make investing in production worthwhile. Okay capitalism, do your stuff.”
Manufacture roof systems, wall panels, Entire kitchens and bathrooms, where labor and factory space make sense. Ship them. Let local skilled crews put everything together.
This is just an example, but it gets us to a different way of thinking about bringing manufacturing back. Not “Which jobs from 1979 can we recreate?”, but “What do we need more of now, and which of those things can we manufacture at scale in places that need good jobs?”
The data business and the Defense Department are already creating enormous demand and watching production follow it. Housing looks like a pretty good candidate to take a hard look at next.
OUTRO
So that’s this week’s Two to the Fifth.
There’s meaning hidden in the algae stuck to your Nobel, an early HIV trial, and besting the Dust Bowl. And sometimes looking across time changes the question. Instead of asking how to bring old manufacturing jobs back, maybe we should ask what America needs more of. And make that.
Thanks for joining us. Now get out of here. Drink responsibly. If you want a cheaper house, manufacture the bathroom like your car. Just skip the sunroof.
And remember: Most things that happen today won’t matter much in the long run. The ones that still matter later are probably worth understanding now.