HigherIQ — Thursday, January 29, 2026

Good morning and welcome to HigherIQ. It's Thursday, January 29th, 2026. Here's a line worth sitting with: "Your zip code is a better predictor of your health than your genetic code." That's from Harvard public health researcher Melody Goodman. And the data backs it up. Two 60-year-old women living just ten miles apart in the Washington DC area face radically different life expectancies: 63 years versus 96 years. Same age, same gender, same country. Thirty-three years of life separating them, determined almost entirely by geography. It sounds like an exaggeration. It's not. And today we're going to dig into exactly why that is, what the healthiest places in America actually have in common, and why the structures around you might matter more than the choices you think you're making. We've also got books being shredded, cars being retired for robots, and a 54-year-old recluse who just quietly overtook Taylor Swift. Let's get into it. A new study from SmartAsset ranked all fifty states on how easy they make it to live healthy, looking at obesity, smoking, exercise, air quality, diabetes, and mental health. The results are fascinating. Utah takes the crown with the lowest smoking rate at 6.9 percent, some of the cleanest air in America, and that absurd 97 percent exercise statistic. Colorado comes second with the nation's lowest obesity rate at 25 percent. Hawaii rounds out the top three, with residents reporting the fewest days feeling physically or mentally unwell. At the bottom? West Virginia ranked worst in five out of ten categories, including the highest obesity at 41 percent, smoking, diabetes, and drug overdose deaths. Here's where it gets interesting. The healthiest states aren't necessarily the wealthiest. Utah's median income is middle of the pack. What they share is access to outdoor activity, strong community health infrastructure, and cultures that normalize movement. Colorado has more trail miles per capita than almost anywhere. Vermont has a "Farm to Plate" program that increased local food production. The takeaway isn't "move to Utah." It's that environment shapes behavior more than willpower does. If your neighborhood has sidewalks, you walk more. If there's a hiking trail ten minutes away, you'll use it. The question worth asking: does where you live make healthy choices easy, or a constant fight against your surroundings? For Atlanta listeners: good news. Atlanta ranked eighth among the fittest cities in America. The BeltLine is doing its job. Speaking of environments shaping outcomes, let's talk about how AI models get trained. A trove of documents just got unsealed in a copyright lawsuit, and they reveal something wild. Anthropic, the company behind Claude, ran something called "Project Panama." The stated goal was to quote "destructively scan all the books in the world." They spent tens of millions buying used books from The Strand and "chronically underfunded" libraries. Then they sliced off the spines with hydraulic cutters, scanned every page, and recycled the paper. Very green. Very thorough. But wait. Before settling on the book-buying approach, an Anthropic co-founder personally downloaded massive collections from pirate libraries. Meta employees reportedly torrented copyrighted books on rented Amazon servers to avoid the activity being traced back to Meta. That's the corporate equivalent of robbing a bank and hiding the cash in your neighbor's garage. Is any of this legal? Maybe. Some judges have ruled that training AI on copyrighted text could be "transformative use." But how you acquire the data matters. Anthropic settled for about 1.5 billion dollars, roughly three thousand per book for half a million works. Statutory damages could have been 150,000 per book. They paid a speeding ticket on their way to a 350 billion dollar valuation. Here's the pattern. Napster built its empire on unlicensed music. Google scanned millions of books first, litigated later. YouTube hosted copyrighted content, then bought legitimacy. In tech, the playbook is: copy first, ask forgiveness later, pay a fraction of your valuation, move on. When you're using AI tools, are you comfortable knowing the sausage factory involves actual book shredding? The S&P 500 briefly crossed 7,000 for the first time ever yesterday before settling back. But the real story was in the earnings. Meta posted record revenue of nearly sixty billion dollars. Microsoft hit 81 billion. Both then announced jaw-dropping AI spending: Meta plans up to 135 billion this year on capital expenditures. Microsoft spent 37.5 billion last quarter alone, up 66 percent from a year ago. Now, 135 billion is a number so large it stops meaning anything. So let's make it concrete. Where does that actually go? Not to software engineers. It goes to physical stuff. Data centers need land, power, water, and massive cooling. Microsoft just signed a deal to restart Three Mile Island, yes the nuclear plant, specifically to power AI. Natural gas demand is spiking because of data centers. Northern Virginia's "data center alley" now consumes more electricity than some small countries. So who benefits besides Nvidia? Utilities, construction companies, REITs that own data center land, energy providers, cooling system manufacturers. There's a shadow economy of AI winners most people aren't watching. If you own an S&P 500 index fund, you're already betting on AI paying off. But the "picks and shovels" play might be infrastructure, not software. One more thing. While pouring money into data centers, these companies are cutting humans. Amazon announced 16,000 layoffs yesterday, 30,000 total since October, roughly ten percent of their white collar workforce. Meanwhile, they're spending 125 billion on AI infrastructure. The math is clear: servers in, people out. When companies say they're "investing in AI," they increasingly mean replacing payroll with processors. Elon Musk announced yesterday that Tesla will stop making the Model S and Model X. Forever. Production ends next quarter. The Fremont factory will be converted to manufacture humanoid robots instead. One million robots per year, if you believe the projections. Let that sink in. The cars that basically invented the modern EV era, that made electric vehicles cool, that made Tesla Tesla, are getting what Musk called an "honorable discharge" so the company can build robots that don't exist yet. The Model S launched in 2012 and proved you could make an EV that didn't look like a golf cart. The Model X followed in 2015 with those falcon wing doors that were either brilliant or ridiculous depending on who you asked. Together, they accounted for only three percent of Tesla's deliveries last year. Financially, not huge. Symbolically? A car company just pivoted to robots. Here's what's worth considering. Musk has said for years that Tesla is really an AI and autonomy company, not a car company. The Cybercab robotaxi doesn't exist in production yet. Optimus isn't on the market yet. But he's betting the factory floor on them anyway. Tesla also reported its first annual revenue decline ever. If Musk is right, Tesla becomes something entirely new. If he's wrong, he just killed his flagship products for vaporware. Either way, 2026 is officially the year a major automaker decided cars are the past and robots are the future. And finally, something delightful. A 54-year-old British electronic musician who once lived in a bank vault and rarely gives interviews now has more monthly listeners on YouTube than Taylor Swift. His name is Aphex Twin. He's clocking 448 million listeners per month. Taylor has 399 million. How? A 24-year-old piano track called "QKThr" from his 2001 album went viral. It's now the background for over four million TikTok videos and half a million YouTube Shorts, soundtracking "subtle foreshadowing" edits and "hopecore" content, whatever that means. This keeps happening. Fleetwood Mac's "Dreams" resurged when a guy skateboarding with Ocean Spray went viral during the pandemic. Kate Bush's "Running Up That Hill" came back after Stranger Things. Social media has become a time machine for music, algorithmically resurrecting songs decades after release. The lesson? Culture doesn't expire anymore. Everything is discoverable. A melancholy piano piece from 2001 can suddenly soundtrack millions of videos about the emotional weight of being alive in 2026. And if you already knew Aphex Twin before this, congratulations on your impeccable taste. Also, check your retinol routine, because we're all getting older together. That's your Thursday briefing. Your zip code shapes your health more than your genes. AI companies are literally shredding books to train models. Big Tech is spending more on cooling systems than most companies spend on everything. Tesla would rather make robots than cars. And a reclusive electronic artist just beat Taylor Swift by accident. The through-line today is environment. The healthiest states didn't get lucky—they built trails and sidewalks. The AI companies didn't just get smart—they built pipelines to acquire data at scale. The algorithms aren't random—they're resurrecting songs based on how we use them. The structures around you are shaping your choices, often without you noticing. Might be worth paying attention to which direction they're pushing. Stay informed, stay curious, and don't be afraid to make a new friend and tell them what you learned today. We'll see you Friday.