A while back I sat down and wrote a letter to the people I love.

It was about artificial intelligence. About AGI, specifically — the kind of machine intelligence that doesn’t just help you write an email but can do most of the cognitive work a person can do. I’d been cornering my family at dinners about it for months, badly, and I finally decided to stop ranting and put my actual thinking on paper. I called it Standing at the Edge of AGI.

I tried to be honest. I tried not to be dramatic.

I failed at the second part, a little. Because when I laid out what I genuinely believed was coming — the quiet, quarter-by-quarter replacement of knowledge work, the layoffs explained away as “restructuring,” careers that stop providing identity or security — it came out dark. It scared the people I sent it to. And if I’m being straight with you, it scared me. That letter became the foundation of this entire site. Everything you read here grew out of it.

Here’s the part I’ve never admitted in public.

The letter had two endings.

The Ending I Stopped Reading

I wrote a bad scenario, the one I just described. And then, because I was trying to be fair, I made myself write a good one — a realistic path where the disruption leads to transformation instead of collapse. Income support that stops being political theory and becomes infrastructure, like roads. Medicine and longevity accelerating inside a single decade. A slow, strange shift where a person’s worth stops being measured by whether they’re needed to survive.

I wrote that part too.

And then I spent the better part of a year acting as if I hadn’t.

I built for the dark ending. I read the dark ending. I watched the layoff numbers roll in and felt the grim satisfaction of being right about the speed. The hopeful pages I’d written sat there like a disclaimer I’d added to cover myself — something I believed in theory and had quietly stopped believing in practice.

Then I read a white paper called Solve Everything, and it walked me back to the half of my own letter I’d abandoned.

The Two Men Who Wrote It

The paper is by Dr. Peter Diamandis and Dr. Alexander Wissner-Gross. If you follow this world at all, you know Diamandis — he founded the XPRIZE and Singularity University, and “abundance” has basically been his religion for two decades. The two of them, along with a rotating cast of sharp people, host a show called Moonshots that I’ve come to look forward to more than is probably healthy.

And then there’s Wissner-Gross.

I need to take a paragraph on this man, because he is a genuine phenomenon. Dr. Wissner-Gross is the closest thing real life has produced to Sheldon Cooper — a physicist and computer scientist whose résumé reads less like a career and more like a clerical error, the kind of person who seems to have collected advanced degrees the way the rest of us collect unread browser tabs. The difference is that Sheldon used his powers to litigate the correct spot on a couch, and Wissner-Gross appears to be using his to help argue that we can cure death and end scarcity inside a decade. I’ll take that trade. When someone that smart tells you the future is bright, you don’t have to believe him — but you do have to sit up.

So I sat up. Then I read the thing twice.

The Claims, and They Are Enormous

Let me be clear about the scale of what these two are arguing, because there is no small version of it.

Their thesis is that the era of artificial superintelligence has effectively begun, and the only real question left is who aims it, and at what. They claim AGI — machines as capable as a median human expert across most valuable work — becomes common and accessible in 2026. This year. They argue that “intelligence” is collapsing into a cheap utility, like electricity, something you pour onto a problem the way you pour concrete into a foundation.

And then they get ambitious.

The core idea is what they call a “solved” domain — a field that tips from needing a stroke of human genius to needing nothing but enough computing power. Their template is AlphaFold: a fifty-year-old problem in biology, the one where a PhD might spend a year mapping the shape of a single protein, that got compressed into a query taking minutes and costing pennies. They argue that what happened to protein folding is about to happen to everything — math, physics, materials science, medicine — each solved domain knocking over the next like dominoes until you get commercially viable fusion, manufactured organs on demand, and a doubling of the healthy human lifespan. By 2035.

They sketch a 2035 they call “the Quiet Hum” — a world where the screaming exponential of the 2020s has settled into something boring and reliable, where the best doctor, tutor, and lawyer on Earth are free utilities replicated at zero cost, and where we stop distributing cash and start distributing capability.

It is a staggering document. It is also, by the authors’ own cheerful admission, written partly by AI and possibly riddled with the kind of confident errors AI is famous for. They put that disclaimer right at the top.

I loved them a little for that.

Because if you’ve read anything else I’ve written here, you know my one religious conviction about working with these tools: verify everything. A confident, beautifully formatted claim is still just a claim, whether it comes from a language model or a brilliant man with a slide deck. So no — I don’t take “fusion and immortality by 2035” as a delivery date. I hold it at arm’s length, the way I hold every dazzling output, and I make it earn its place.

But that’s not the part that got me. The part that got me was quieter.

Why It Actually Landed

Here’s the thing I wasn’t ready for.

I’d written my hopeful scenario as a wish. A thing I hoped someone, somewhere, would have the decency to build. Diamandis and Wissner-Gross wrote the same future as an engineering project — with mechanisms, incentives, and a build order. They gave the bright ending an engine.

And the overlap with my own letter was genuinely eerie to read.

I had written that income support in an AI economy “isn’t charity — it’s core infrastructure, like roads or electricity.” They argue for almost exactly that, dressed up as “Universal Basic Capability.” I had written that the deepest change coming isn’t technological but human — that we’d have to find worth somewhere other than being needed for our labor. They land on the same shift, calling the people of 2035 “Creators of Meaning.”

I reached those conclusions by staring into the dark and forcing myself to imagine a way out. They reached them by staring at exponential curves and getting excited. Two completely different doors. Same room.

That’s what moved me. Not that two famous optimists are optimistic — that’s their whole brand. It’s that an honest, frightened look at this technology and a giddy, ambitious one can arrive at the same hopeful place. The good scenario isn’t the naïve one. It might be the engineered one.

I’d been treating my hope like a polite afterthought. They reminded me it was a blueprint.

What Changed, and What Didn’t

I want to be careful here, because the easy move is to swing from doom to hype, and I’m not doing that. The dark scenario in my letter is still real. The layoffs on this site are still real. None of that gets erased by a beautiful PDF.

What changed is that I stopped reading only one of my two endings.

I now believe both scenarios are genuinely live — the brittle, concentrated, bureaucratic version where superintelligence mostly optimizes ad clicks, and the abundant version where it cures things. The white paper’s whole argument is that the difference between those two futures isn’t decided by the technology. It’s decided by what we point it at, and by whether ordinary people understand the stakes early enough to push.

Which, I realized, is the entire reason I started writing in the first place.

The people who navigate this best won’t be the ones who picked the right prophecy. They’ll be the ones who saw both endings clearly, prepared for the hard one, and refused to stop building toward the good one. That’s not optimism. It’s not doom. It’s the posture I should have held the whole time and didn’t.

What I’d Tell You If We Were Having Coffee

Read the white paper. Read it the way I did — twice, with your skepticism fully switched on. Roll your eyes at the parts that read like a science-fiction press release. They earned a few eye-rolls.

But don’t stop there, because that’s the mistake I made with my own letter. Don’t read only the half that confirms what you already feel. If you’re a worrier like me, make yourself sit with the bright version until it stops feeling like a fairy tale and starts feeling like a thing that could be built — by people, on purpose, including you. And if you’re a natural optimist, do the opposite: sit with what gets broken on the way there, because plenty does, and the people in the rubble are real.

The future isn’t a forecast you wait to come true. It’s a fork, and we’re standing at it right now.

I wrote a letter about the edge of AGI and accidentally wrote two endings. For a year I read one of them. A couple of brilliant strangers handed me the other one back and said, essentially: this part is real too — now help build it.

So that’s what I’m doing. Awake, honest, and — for the first time in a while — actually hopeful.

You’re welcome to read over my shoulder.

Continue reading: Read the pillar — Your Income in the AI Era

This article is for informational purposes only and is not financial advice. Consult a qualified professional for personalized guidance.