AI: The Wild Ride and the Future of Work
What we all know about AI, and what we’re yet to learn
- The stock‑market buzz: For the last three years, an AI‑driven bonanza has been pushing prices to the moon, even when the market takes a few fatalistic dips.
- The job‑shuffling wave: AI rolled out its tech toolkit, promising to reshape the global landscape—and it might even politely shove out a handful of roles that used to demand training and technical know‑how.
So if you’re looking to ride the AI wave, keep your hands on the wheel and your eyes on the horizon. It’s a thrilling tour of optimism and, occasionally, a reminder that the future belongs to ones who can adapt—especially when it’s on autopilot.
AI’s New Era: Stocks, Jobs, and the Smart Move to Cash
It’s a wild ride these days – AI’s been catapulting the market into record highs, but the vibe is unmistakably that of a super‑bubble. The crash is a ticking time‑bomb that could drop the market by 50% or more, and it might happen sooner than we think. But shorting the big indices now? That’s a risky gamble – you could lose a ton if the market doesn’t slide.
Why “Get Your Cash Out of the Card” Is the Better Strategy
- Let the bubble inflate. The rise isn’t coming to an abrupt end. It’s the kind of bubble that lingers longer than most people expect.
- Shorting is a double‑edged sword. If you bet against the market and it turns out to go up, you’ll be in for a nasty loss.
- Trim your equity. Reduce your stock allocation and pile more into cash. That way, if the crash does hit, your portfolio is cushioned.
AI: The New Job‑Altering Ally
Sure, AI will delete or make some roles easier to replace. That’s the nature of any tech revolution – old jobs sell out, new ones rise. But even in a world where computers outshine humans at basic math and reading, teachers aren’t going anywhere. They’re just swapping the routine drills for teaching critical thinking and reasoning, skills that the algorithms simply can’t nail.
In short, the changes across the board will feel giant, but they’re still manageably incremental. The world is shifting, not tearing itself apart. Just keep a lid on your equity, load up on cash, and look forward – because the stock market’s superstasis is a bit like a rollercoaster that’s still on the track.
The Limitations
AI’s Power‑hungry Side‑Story
Artificial Intelligence is dazzling the world, but that glow comes at a hefty price tag—talk about a power hungry beast that’s not going to stop for a coffee break.
Why the Energy Guzzler
- Processing Power: AI chips are faster than a caffeinated rabbit, but every new sprint consumes a boatload of juice.
- Training Sets: Feeding AI more data is like stuffing a hungry dinosaur; the more you feed, the more it grows hungry.
- Electricity: Big data centers cram thousands of these power-hungry chips together, turning the grid into a giant battery.
Enter the Megawatt Jungle
The current generation of semiconductors is like a rollercoaster: zooming faster yet screaming louder. Soon, the next wave of chips will arrive, but they’re no small fry—they’re “mega‑processors,” demanding mountains of amps.
Nuclear Power: The Unexpected Hero?
Some folks are flipping the script and suggesting nuclear reactors as the ultimate power plug. Small Mod “Micro‑Pods” or gigantic “Smith’s Sisters” could keep the AI engines humming, especially for those massive, neon‑lit data centers.
The Egg‑On‑A‑Stick Demand Curve
Here’s the kicker: AI’s energy needs curve isn’t linear. Think of it as a “grow‑fast” demand—every little upgrade bites off a big chunk of electricity. The result? AI is hitting the practical ceiling of performance before the next chip can scramble up the next level.
Energy: The Real Winner
If AI were a race, the real champion would be energy. The “AI race” turns into an “energy race.” And as it plays out, the two big players—US and Russia—are sprinting ahead. Green‑light check: China’s battery stays near Russia for its gas supply, and Europe’s wheels are turned by the US‑Russia combo.
Sanctions & the Irony Loop
Think banning Russian oil is a good idea for Europe? Surprise! When Russia can haul its gas home, it can stash it like a giant treasure chest. That stash fuels AI models and crypto mining, giving Russia an unintended edge.
What Does It Leave for Others?
For short‑sighted Europeans juggling politics and a resource‑tight China, it’s a classic case of “the wise man is not always the winner.” If you’re not careful, you’ll end up caught in this energy ring‑race, chasing a ghost that’s sitting pretty on a nuclear reactor.
Bottom line? AI isn’t just about pushing the limits of code. It’s now also a power‑monopoly clout battle. If you want to be a player in this game, knowing how batteries (or reactors) spin is your new secret weapon.
AI Lacks Common Sense
What AI Can’t Really Do (And Why Humans Still Rule)
The Unspoken Bouncer: Conservation of Information
Imagine a magic spell that keeps the universe from spitting out new magic at will. That spell is the Law of Conservation of Information in Search. It’s a no‑nonsense, math‑proof‑backed rule that tells us AI is super fast at finding the stuff that already exists, but cannot magically invent fresh facts. The real treasure lies in the human side of things—creativity, art, stories, and original riffing.
When AI Becomes the Echo Chamber
On the internet we’re already drowning in a deluge of AI chatter. The trick is simple: once an AI adds its own “grand synth” to the training mix, that noise becomes part of the next generation’s “facts.” Unfortunately, bots love to hallucinate—confabulations that look oddly convincing but are completely hollow. So every time you pile AI output onto more AI output, the quality of the learning dataset shrinks, and new models by the same logic get even thinner.
- Experts ≠ Light Work: Only a subject‑matter sleuth can sift these smears and keep the data solid.
- Costly Curators: Because you need a professor’s eye, the whole “AI one‑click” dream gets a real‑world break.
The ‘AI Gravatars’ Missing Common Sense
A recent showdown between an AI and a bunch of 3‑ to‑7‑year‑old kids showed just how clueless the machine can be. The task? Draw a perfect circle using whatever tools were handed out: a ruler, a teapot, and an oddball like a stove.
The AI tried to be clever, treating the ruler as a drawing instrument and pulling out a fancy “drafting circle.” That plan? Futile. Meanwhile, the kids saw the teapot’s lid—simple, round, real—and traced it to beat the computer. No fancy math, just good old practical sense.
Industry Power Struggle: Big Name vs. Big Pass
Tech titans like Microsoft and Google are discovering that a clever shortcut outclasses their grand designs: folks swipe fresh AI output (the so‑called “big ticket’’) and plug it straight into a new model. It’s like using the latest blockbuster as a training soap. Costs? Low. Gains? Surprisingly high.
Just so you know, the “stealing” in this context isn’t a serious theft; it’s more of an unfrequent copy‑paste before the law catches up.
What Does This Mean for the Future?
The headline is candid: AI’s sky‑high profit numbers are crashing. Billions poured into AI labs may not pay off as promised. But the machine’s core capabilities—finding, connecting, not creating—still survive. The bottom line: humans keep the creative spark alive; AI is just a super‑fast librarian.
Sam Altman: Innovator or Salesman?
Sam Altman: The AI Hotshot Who Keeps the Lights On
When you think of the AI world, one name lights up the room: Sam Altman. He’s the big cheese behind OpenAI, the brain behind the hit ChatGPT app. A quick trip down AI history shows a wild ride:
- 1950s: AI births
- 1980s: The dreaded AI Winter
- 1990s–2000s: A quiet lull
- 2010s–present: Comeback saga
ChatGPT became the most‑downloaded app ever in just a few months, and today it can’t even begin to count its hundreds of millions of users.
The Boardroom Brawl
Last year, the OpenAI board fired Altman because the company was supposed to stay non‑profit and focus on AI for the common good. But Altman had grander dreams: turning the venture into a for‑profit wheel‑and‑deal that would eventually spark a multi‑hundred‑billion‑dollar IPO.
When the top engineers threatened to quit and follow him to a new startup, the board had a change of heart and brought him back. This new twist re‑establishes Altman as the board’s main player, though the legal details are still shrouded in mystery.
Superintelligence: Straight Up Vision
Altman keeps talking about superintelligence, also called advanced general intelligence (AGI). He says it’s “general” enough to think like humans, but way better.
Take the ape‑human analogy: imagine humans as the “smart” apes relative to the computer masters. Altman claims:
“ChatGPT is already more powerful than any human ever existed.”
He also predicts AI will:
- do real cognitive work by 2025
- uncover novel insights by 2026
Reality Check: The Limits of AI
These bold claims simplify a complex truth. Quickly, training data gets piled up with earlier AI output, making new models paradoxically less clever.
Mathematics also backs the Law of Conservation of Information—computers can locate information faster than humans but can’t create new facts. In plain language: “No‑no, they’re not truly thinking, just faster at making connections.”
Apple’s Hot Take
A recent Apple paper reveals a deep insight:
“Across numerous puzzles, frontier LRMs collapse in accuracy beyond certain complexities. They plateau and then fall off as problem difficulty climbs, even when the token budget is ample.”
This shows that beyond a certain point, computational brute force can’t push past logic limits.
Why Superintelligence Won’t Arise
Developers still can’t code abductive logic—that’s the gut feeling or common sense you apply to puzzles. That’s arguably the smartest part of human reasoning.
At the end of the day, the “superintelligence” hype appears to be just another Silicon Valley pitch. Altman’s future grand plans may spice up the headlines, but the path to true AI cognition remains a steep climb.