Welcome to the weekend.

Prime Numbers

1,000,000,000 — The federal government is providing Constellation Energy a $1 billion loan to restart Three Mile Island Unit 1 (now renamed Crane Clean Energy Center) by 2027, covering most of the $1.6 billion project cost to power Microsoft’s data centers and address soaring electricity demand in a region serving 65 million people.

70 — An analysis of 328,744 publicly shared ChatGPT messages revealed that 70% contained at least one emoji, with the chatbot using checkmarks 11 times more often than humans do—a telltale sign for spotting AI-generated text in Turing tests.

61 — The share of 12th grade girls wanting to get married someday plummeted from 83% in 1993 to 61% in 2023, while boys’ views remained virtually unchanged, creating a new gender gap where boys are now more marriage-minded than girls.

Technology Defeats Cancer

Most people still think of cancer as a death sentence requiring brutal chemotherapy, but the war has fundamentally shifted from carpet-bombing to precision strikes. Malignancies like advanced melanoma and non-small-cell lung cancer—once guaranteed killers—are now curable because oncologists have moved beyond poisoning the entire body to targeting cancer with pharmaceutical precision. James Watson’s seemingly impossible goal of flipping cancer statistics isn’t fantasy anymore: we’re approaching a reality where only 100,000 die instead of only 100,000 surviving. This transformation means the diseases that terrified previous generations into paralysis are becoming manageable conditions in yours, fundamentally changing how you should think about prevention, early detection, and treatment decisions when cancer touches your life or business. WSJ (6 minutes)

Intelligence’s Fatal Flaw

We’ve spent a century measuring intelligence as if it’s one universal problem-solving engine, yet people who ace IQ tests are no happier than those who fail them—and across 50 years of data, higher scorers are actually slightly less happy. The reason reveals a fundamental misconception: what we call “general intelligence” only measures skill at well-defined problems with clear rules, stable relationships, and indisputable answers—math tests, chess games, vocabulary matching. But life’s most consequential challenges are poorly defined: “How do I live a meaningful life?” “Should I change careers?” “Why can’t I find love?”—problems where the rules shift, boundaries blur, and solutions that worked at 21 fail at 31. This distinction explains why brilliant physicists can believe absurd conspiracies, why elite professors commit career-ending blunders, and why a century of solving well-defined problems (eradicating polio, landing on the moon, gaining 15 IQ points) hasn’t budged human happiness one bit. When you spend your life optimizing for test scores and promotions—treating every challenge as if it has a correct answer you can calculate—you’re wielding the wrong tools entirely, like trying to fix a marriage with calculus. The grandmother who can’t work her TV remote but knows how to raise good people and carry on through tragedy possesses the intelligence that actually determines whether your one finite life feels worth living. Experimental History (7 minutes)

Data Centers on the Moon

The world’s richest men are seriously discussing building data centers on the moon—not because the economics make sense today, but because AI’s energy appetite has grown so ravenous that outer space suddenly seems more practical than dealing with Earth’s power grid constraints and regulations. Solar panels in space harvest constant, direct sunlight without clouds or nighttime interruptions; cooling happens naturally in the vacuum; and crucially, there are no pesky permits or neighborhood objections slowing construction. Musk envisions lunar bases catapulting satellites that collectively generate 100,000 gigawatts annually—power so absurd that “Back to the Future” used just 1.21 gigawatts as the threshold for time travel. When billionaires start treating moon colonization as the more realistic path than fixing terrestrial infrastructure, you’re witnessing either visionary ambition or the final stages of a speculative bubble where no amount of energy feels sufficient. The AI race has become so untethered from present-day economics that proposals involving mass drivers and orbital data farms receive earnest consideration—because admitting you might need to slow down or optimize differently would mean conceding competitive advantage. What began as “we need more power plants” has escalated to “we need to industrialize the solar system,” revealing how thoroughly the AI arms race has warped our sense of proportionate response to computational demands. WSJ (4 minutes)

Your Customer’s New Boss

Retailers spent decades fighting to own the customer relationship, yet AI agents are quietly becoming the new intermediary—and consumers might prefer it that way, with 30-45% already using AI for product research and ChatGPT shopping referrals growing sevenfold in just one year. The existential threat isn’t that agents make recommendations; it’s that they could reduce retailers to “dumb fulfillment pipes,” controlling transactions, compressing margins, and capturing all the behavioral data while brands become unknowing participants in someone else’s marketplace. Shoppers currently trust retailer-owned agents three times more than third-party ones, but that trust gap is temporary—a narrow window before AI platforms like ChatGPT become the default starting point for all commerce. The retailers who survive won’t be those with the best products or prices, but those who make shoppers care about where they buy, not just what they get. This means weaponizing exclusivity (products, bundles, loyalty multipliers unavailable through agents), reinventing retail media around metadata and prompt data instead of search keywords, and fighting to retain control over checkout and customer data even when partnering with AI platforms. The uncomfortable reality: AI shifts loyalty from brands to outcomes, and if your unique value isn’t apparent to both humans and algorithms, you’ll wake up one day to discover you’ve been quietly demoted from retailer to warehouse—still doing the work, but no longer capturing the profit or relationship. Bain & Company (6 minutes)

The AI Pioneer’s Paradox

The man who invented the core components of modern AI—and was proven right for four decades—now says the entire industry is chasing a dead end with ChatGPT-style language models. Yann LeCun, one of AI’s “godfathers” and winner of computer science’s highest honor, compares today’s most advanced LLMs to a cat’s mind, and he thinks the cat is smarter. His conviction: true machine intelligence will come from “world models” that learn by observing like babies do, not by predicting text from massive databases—a view so contrarian that Meta sidelined him, gave him a 28-year-old boss, and cut resources to his division even as they pour billions into the LLM race. If the guy who was right when everyone else was wrong is now saying everyone is wrong again, the billions flowing into scaling up ChatGPT competitors might be the most expensive bet in tech history—and your AI strategy may need a complete rethink before his three-to-five-year prediction comes true. WSJ (7 minutes)

Toddlers: Genius Idiots

You can experience the purest love imaginable when your toddler hugs you and simultaneously conclude you’ve spent three hours with someone whose IQ is 20—and both feelings are completely valid. Toddlers are paradox machines: they memorize entire books on first read but can’t find Curious George staring directly at them; they learn Mandarin by osmosis while thinking “you” is their own name; they’re comedic geniuses who find “kerplunk” devastatingly funny for ten straight minutes. The humbling truth is that toddler parenting matters far less than our obsessive opinions suggest—every strong parenting conviction has an equally strong opposite, and everyone’s silently judging everyone else’s choices while secretly having no idea if they’re doing it right. Instead of trying to shape these tiny dictators into something they’re not, your real job is helping them become the best version of who they already are, which means accepting you’ll mess up constantly and mostly just trying to have fun with your brilliantly stupid little companion. Wait But Why (6 minutes)

Brevity Requires More Work

Everyone thinks being succinct means using fewer words, but the executive who can’t focus on your rambling presentation doesn’t need shorter sentences—they need you to have done more thinking before you opened your mouth. “Be Brief, Be Brilliant, Be Gone” isn’t about word count or clever frameworks; it’s about having such deep contextual mastery that you know exactly which details matter to the company’s broader strategy and which are just noise you find personally interesting. The paradox: becoming succinct requires significantly more preparation than being comprehensive, because you must gather context from across the organization, crystallize one main point from the chaos, and then fight your own discomfort with leaving things unsaid. If you’re rambling in meetings or watching executives’ eyes glaze over, you haven’t done enough work beforehand—and that unpreparedness is costing you credibility with the people who control your career trajectory every time you fail to connect your work to what actually matters. The Uncommon Executive (5 minutes)

Should We Work Together?

Hi! I’m Kyle. This newsletter is my passion project. When I’m not writing, I run a law firm that helps startups move fast without breaking things. Most founders want a trusted legal partner, but they hate surprise legal bills. At Westaway, we take care of your startup’s legal needs for a flat, monthly fee so you can control your costs and focus on scaling your business. If you’re interested, let’s jump on a call to see if you’re a good fit for the firm. Click here to schedule a one-on-one call with me.

Weekend Wisdom

Wine is constant proof that God loves us and loves to see us happy. - Benjamin Franklin

Weekend Briefing

Weekend Briefing No. 615