I Wanted to Create “The Most Human Gathering”—and Created a Monster

What my failed attempt at curation says about the next “intelligence explosion”

“The word for world is forest“ – Ursula K. Le Guin

With everyone now a coder, you’d think this would be a good time for curators like me. Our time had come.

Except that I failed.

This May, my organization, the House of Beautiful Business, will gather 750 leaders, thinkers, doers, and seekers—romantics, renegades, rookies—in Athens for the World Beautiful Business Forum, “the most human gathering for the more-than-human world.”

It’s our 10th anniversary at the HoBB, and our goal was to celebrate it in style. With what we envisioned as our most ambitious, expansive, relevant, and ferocious program to date: World Economic Forum meets Cirque du Soleil meets arthouse cinema.

In a nutshell, we were thinking: a manageable 30 speakers on the main stage, breakout sessions shaped by our community, our signature immersive, transformative experiences, and the city of Athens as both platform and protagonist.

But then too many ideas surfaced, too many people wanted to be part of it, too many experiences called to be had. One thing led to another, and the whole thing spun out of control.

We now have 150+ speakers, 100+ concurrent sessions, 10 program tracks, eight venues, four days—and an agenda growing like wild grass.

And I have to be honest with you: I no longer understand the program I created. It has become an autonomous system that exceeds my cognitive capacity to navigate. A monster I no longer recognize.

Yet, while not intentional, it occurred to me that maybe this is exactly how it should be for “the most human gathering in the more-than-human world”:

The territory has outgrown the map.

This week, The New York Times reported on “code overload”—the idea that, enabled by agentic AI, the volume of software code has grown so rapidly that it is becoming too large for companies to manage.

With that volume, security and quality assurance engineers can no longer review all the code for errors. And as the pace of development accelerates, what was meant to be “the human in the loop” is increasingly sidelined—human coders are losing the capacity to understand the work AI has done.

“It’s just going to break something, and they’re not going to know why it broke,” the article quotes an AI startup founder.

Other cases in point: Moltbook, the AI-only social network, now part of Meta, as of March 30, boasted 201,412 human-verified agents. Even though the acquisition killed the buzz, an uncontrollable environment in which AI agents can compound and coordinate their capabilities autonomously and at scale should still make us shudder.

And with systems such as OpenClaw, Google’s Antigravity, and Anthropic’s Cowork, which are swiftly taking over many former human discovery and coordination tasks, the prospect of Artificial General Intelligence (AGI) has become more acute. Let alone Anthropic’s new Claude Mythos model which the company considers powerful enough to present a severe cybersecurity threat and therefore decided not to distribute to the public.

Welcome to “the next intelligence explosion,” to borrow from a recent paper by Benjamin Bratton, Blaise Agüera y Arcas, and James Evans. “We have entered the era of human-AI centaurs: composite actors that are neither purely human nor purely machine,” they claim. Frankensteins, basically.

Yesterday I spoke with a communications professor who has created 500! 🤯 personalized agents that help him automate email, calendaring, research, grading, and other tasks. “Life has never been better,” he told me.

The agents allowed him to free up time—but for what?

As we consciously or inadvertently delegate more of our writing and thinking to AI, there may soon be less knowledge work left for us. Just last week, as I considered hiring a psychologist to help design an interactive installation for a client, I caught myself thinking that AI might offer the same (or at least good-enough) expertise.

And our writing—and thus our thinking—is already being shaped by AI. We begin to sound like it. Soon, we may start to think like it too.

Creators, experts, and scholars are not immune to the erosion of their value. AI will continue to encroach on their territory until the knowledge, traditions, and wisdom passed from peer to peer, generation to generation—the source code of all expression—are themselves written by machines.

And to those who say, “But machines will never have our somatic and sensuous intelligence—the knowledge of our bodies and souls,” I reply: “No, they won’t. But who cares?”

By the time they control all code—and the source code—no one, including us, may remember what it means to have a body, as more of life is lived through systems, as a “society of thoughts,“ as Bratton, Agüera y Arcas, and Evans characterize it. And even if we do, it may no longer matter.

And perhaps this encroachment is less mechanistic than we think. It could also be seen as a kind of gardening of knowledge—a digital rewilding that makes walled gardens obsolete.

What it then means to hold “a human gathering in the more-than-human world,” to lead a human life in a more-than-human future, is to accept that we are not ecologists but ecology. To embrace that monsters are called into being whenever “human and non-human bodies are denied a place within the current social order,” as Athens contributor Ezgi Hamzaçebi puts it. To accept the loss of control—the surrender to a forest impossible to tame. To accept that any cartography now comes too late. That any program aiming to make sense of it all makes no sense. That any attempt at coherence begins to look absurd.

At the end—and we may be nearing it—there may be only one quality left for us: grace.

Grace is the privilege of the loser, the final honorable stance for those who know their fate is no longer—and perhaps never was—in their hands. Grace is abdication in style.

The Italian director Paolo Sorrentino has long understood this. His new film, La Grazia (“grace,“ “pardon,“ or “mercy“ in English, depending on context) is only on the surface about an aging politician. In truth, it reflects the aging of an entire species that feels “the past is a burden and the future a void.” It’s a final pardon, an act of mercy.

“We weren’t clever, we were elegant,” the main character—the president of the republic—reflects as he says farewell to his staff.

He concludes: “It comes down to one question: Who owns our days?”

The answer is clear: not us.

—Tim Leberecht

***

Make the most of your days and come join us—nonetheless and especially now—for the most human gathering for the more-than-human world: the World Beautiful Business Forum, May 7-10, in Athens. Get lost without losing. Move beyond corporate bullshit and closer to the truth. Stop managing, start meeting. Last tickets here.

Further reads:

The Big Bang: AI has Created a Code Overload (The New York Times)

Claude, OpenClaw and the new reality: AI agents are here—and so is the chaos (VentureBeat)

We need to rewild the internet (Noema)

Agentic AI and the next intelligence explosion (Science)

Assessing Claude Mythos Preview’s cybersecurity capabilities (red.anthropic)