Envisioning

What would you do with unlimited resources?

Most of us spend our lives managing scarcity—scarcity of money, time, and attention. We occasionally manage to reprioritize and act in our best interests, but for the most part, life is about trying to get the most out of limited inputs. Whether you’re putting extra hours into a project or budgeting time to improve yourself, everything comes down to trade-offs. You get 24 hours each day, and unless you’re unusually rich, you treat money and time as finite. You learn to make them stretch.

But what if those limits vanished? What if you had access to an intelligence that could think faster, deeper, and more collaboratively than any human? What if you could deploy it however you liked, working tirelessly towards whatever you set your sights on? What would change about how you spent your time, your attention, your life?

AI is here, and it’s reshaping how we think about what’s possible. But here’s the tricky part: we judge AI based on our limited, human experience. If an AI nails a complex task, we’re impressed. When it fails, especially in ways that seem nonsensical, it’s jarring. Sometimes it fails at something so simple, it leaves you wondering how it ever got anything right in the first place. Reload the same query, and you might get the correct response the next time. But the inconsistency sticks with you.

The challenge is that AI systems don’t operate by our logic, and so their strengths and weaknesses don’t map cleanly to our expectations. AI is like a tool with invisible edges—it can carve out remarkable capabilities in one direction, while creating blind spots in others. And it’s not always about the AI’s inherent quirks, but about who is shaping these systems and for what purpose. The decisions about how our tools are designed and deployed are being made by a small group of people in a handful of organizations. They’re deciding the future for all of us.

That’s why staying informed matters. To understand what AI can do for you takes work. It takes making mistakes and adjusting your assumptions. Learning to interact with these systems isn’t a passive process. It’s active and humbling. You have to figure out when the AI is really listening to what you mean versus just giving you a plausible answer. It’s about learning the language of these models and figuring out what you can ask of them.

In a world where intelligence feels infinite, the challenge isn’t getting more of it. It’s knowing how to use it.

—

As always, if this content resonates and you want to meet fellow readers, make sure to join our amazing WhatsApp community.

Until next week,
MZ

The true story of The Limits to Growth (120 min)

The Tipping Point Podcast is adjacent to AI, but fundamental for system thinkers and concerned citizens everywhere and was researched, created, and produced by Katy Shields and Vegard Beyer. I can't recommend it enough.

Tools for understanding things (35 min)

The NYT's Hard Fork podcast interviews author Steven Johnson about Google's new NotebookLM platform. All about Centaur thinking - using AI for augmented collaboration. I had heard about Johnson at Google but had no idea he was involved in building this product. First time I hear about "AI Vertigo" which sounds a lot like future shock. Here's the Chicken PDF Johnson mentions around the 29 min mark.

Sam Altman on How I Write (50 min)

Pretty interesting interview with Sam Altman about how he works, thinks and writes. Not an AI deep dive but insightful into how someone deeply involved with AI actually uses it on a daily basis. (I'm also a fan of the .38 Muji pens)

The Intelligence Age

Sam Altman is catching a lot of (justified) attention these weeks, with the apparent changes going on with OpenAI's corporate structure. Nonetheless, I find him a fascinating thinker when it comes to the potential implications of AGI and always worth a read. This essay is no exception.

It is possible that we will have superintelligence in a few thousand days (!); it may take longer, but I’m confident we’ll get there.

Quick links

Cautiously optimistic about this marketplace initiative by Cloudflare to compensate "content creators" for their training data.
Dan Shipper from Every did what I've been wanting to do: a 30 min tutorial on how to build an app with Cursor to scratch your own itch (tho I'd still use Claude 3.5 over O1).
In a similar vein: building with LLMs has never been easier. Here is another cool hands-on experience article.
From Reddit, a duet with the advanced voice model.

Accurate

Bill Gates on AI, Wealth, and the Future (6 min)

My favorite talk show host talking with Bill Gates about AI's future, wealth inequality, and his charitable pledges

It is the first technology that has no limit.

Cool generative AI experience - turns a photo into a unique poem, based on a number of inputs and variables. By designer Justus Bruns.

You are receiving this newsletter because you signed up on envisioning.io or Substack.

The Next 1.000 Days

More from Artificial Insights

Newsletter