Intelligence in the Age of AI Agents

The year was 2011. We called AI “machine learning” back then, and I was working to grow my hedge fund I’d recently launched with 3 other partners. We used python to scrape financial statements + pull them from EDGAR and then we performed sentiment analysis alongside more traditional financial analysis. Fund failed, but I fell in love with the leverage tech created and the new capabilities it made possible.

So I pivoted to tech investing.

I saw a lot of deals (2,000+ pitch decks, 800+ pitches) and invested in a dozen companies. Went on diligence visits to see close to 50 companies. I remember one of my first office visits. This company was run by two scary smart programmers out of a space in Brooklyn that smelled permanently of pizza wafting from the slice shop below. Their great innovation was a novel chess engine they were applying to business problems in unique ways. It was brilliant, of course. It could thrash a grandmaster without breaking a sweat, calculating 3 billion moves a second while its server fans hummed a tune of quiet superiority. They were using this mind to see business moves instead of chess ones.

Back then, our big idea for human and machine collaboration was what the chess nerds called a “Centaur.” You’d take one grandmaster, a person with all the intuition and strategic flair that comes from being a lumpy, biological creature, and you’d pair them with a top-tier chess engine. The human would handle the grand strategy, the unknowable art of the game. The machine would handle the brute force calculation, checking every move for blunders.

Together, they were unbeatable. Better than any human alone, and better than any computer alone. For a decade, this was the model. The human as the wise rider, the AI as the tireless horse.

It’s a lovely image. It is also, I must warn you, about as current as a steam ship.

From my perch as a fellow who invests in this sort of thing, I get to see the future arriving in bits and pieces, often in messy beta tests and bewildering technical papers. And the future isn’t a Centaur. It’s something vastly more complex, more powerful, and frankly, more interesting.

We are living at the dawn of the Centaur Swarm. And its making me thing an awful lot about the nature of intelligence itself.

I’ve captured my thoughts on this topic before:

But my thoughts have evolved after seeing the most recent technology.

To understand what that means, you first have to appreciate that we’ve been thinking about intelligence all wrong. We thought of it as one thing, a single scale of pure processing power. More horsepower in the brain meant more intelligence. But intelligence isn’t a single engine. It’s more like a chaotic, bustling kitchen full of specialists.

There’s a pattern recognition chef, a short-order cook for abstract reasoning, a sous chef for adapting to new goals, and a head waiter who handles all the social and emotional intelligence. A truly smart mind isn’t just one brilliant cook. It’s a perfectly coordinated kitchen staff.

Since 1956 and perhaps even before then we have tried to build one single AI that could do it all. The master chef. The results were impressive but always a little… lopsided. An AI could write a perfect sonnet but couldn’t figure out that a cat is not a suitable hat. It could identify a tumor in a scan with superhuman accuracy but couldn’t tell you if the patient was feeling sad about it.

So, the really clever people stopped trying to build a single master chef. Instead, they started building the whole kitchen staff.

This brings us to the new workforce.

At the heart of it is a creature called an AI agent. Now, “agent” is one of those Silicon Valley words that can mean anything and nothing. So let’s be simple. An AI agent is a tiny, self-contained specialist. It is a computer program with one very specific job, a goal it pursues with the relentless, unblinking focus of a hummingbird.

Imagine you could hire an employee who needed no sleep, no pay, no coffee breaks, and whose entire existence was dedicated to one task. Say, finding the absolute cheapest flight to Honolulu for the third week of June, considering weather patterns, airline reliability, and your known preference for an aisle seat. That’s an agent. You give it a goal. It goes off and does it.

You probably need to arm it with tools. There are a growing bank of tools that have been open-sourced. Every tool you add to this agent’s toolbox will increase its capabilities.

It’s a useful little thing on its own. Perhaps you can power it up with a few additional tools… but its real power comes when you stop hiring just one.

You hire ten. Or ten thousand.

This is the swarm. It’s a multi-agent system where all these little specialists can talk to each other, delegate tasks, and collaborate to achieve a goal far too complex for any single one of them. And this is where the Centaur model gets its spectacular upgrade.

The new system isn’t a rider and a horse.

It’s a three-tiered command structure, and it looks something like this:

Tier 1: Human Strategist. At the apex is the human, who provides the high-level, creative, and often ambiguous intent. They set the ultimate goal without needing to define the specific steps.

Tier 2: Master AI (The Orchestrator). This central AI acts as the bridge between intent and action. It receives the human’s vision and deconstructs it into a logical, structured plan composed of thousands of precise sub-tasks. It then assigns these tasks to the appropriate agents.

Tier 3: The Swarm. This is a dynamic, large-scale workforce of specialized AI agents. Each agent is created to perform a specific, narrow function. They execute their assigned tasks efficiently and then cease to exist, making the system incredibly agile and scalable.

At the very top, you have the human.

Let’s call her the Human Strategist.

Her job is no longer to do the work, or even to guide the work second by second. Her job is to provide the one thing a machine cannot.

Intent.

She sets the grand vision. She defines the mission. She asks the big, fuzzy, wonderfully human question: “What if we tried to do this?”

Below her is the second tier, the Master AI, or the Orchestrator.

Think of this as the world’s most efficient and terrifyingly competent project manager. It takes the human’s grand, poetic vision and breaks it down into a thousand precise, logical, actionable sub-tasks. It understands the goal, it knows its team, and it handles all the assignments.

And beneath the Orchestrator is the third tier. The swarm itself. A teeming, buzzing ecosystem of specialist agents, spun into existence for a single purpose and vanishing the moment their job is done. They are the researchers, the coders, the analysts, the designers, the marketers. They are the new workforce.

This structure creates a level of cognitive leverage that is frankly difficult to comprehend. It allows a single human mind to direct the productive output of what would have once been a massive corporation.

Let me show you what I mean.

A good friend of mine, a brilliant biologist, is trying to cure a particularly nasty type of brain tumor. For years, her process was painfully slow. Read hundreds of papers. Form a hypothesis. Spend months in the lab testing it. See it fail. Repeat. Her genius was bottlenecked by the sheer, grinding labor of it all.

Yes she had a team. Yes she had resources. But the size of the problem made those resources insufficient.

Today, her lab looks very different.

Last week, she sat down at her terminal with a cup of tea and a new, audacious goal. She wanted to explore a wild theory connecting the tumor’s growth to a specific metabolic pathway that most of the literature had dismissed.

She typed her intent into her system. A simple, human sentence. “Explore all non-obvious connections between glioblastoma growth and mitochondrial protein expression, and propose three novel, testable drug compounds that interrupt that process.”

Her Orchestrator AI, which she’s nicknamed “Archimedes” because she’s a big geek like me immediately went to work. It didn’t just understand the words. It understood the intent. It understood what a “non-obvious connection” meant. It understood what “testable” implied for a drug compound.

Archimedes instantly deployed a swarm.

First, a team of ResearchAgents was born. Imagine a thousand tireless librarians who can read every medical paper ever published, in every language, simultaneously. They inhaled decades of research in about ninety seconds, flagging not just the obvious papers on glioblastoma, but papers on cell metabolism, mitochondrial function, and obscure yeast genetics that happened to mention a relevant protein.

Then, Archimedes spun up the HypothesisAgents. These are the creative ones. Their job is to find connections nobody has seen before. They took the mountain of data from the researchers and started building bridges. They found a faint link here, an ignored correlation there. They generated over ten thousand potential hypotheses, most of them junk, but a few of them utterly brilliant.

Archimedes filtered these hypotheses, discarding the ones that were biologically impossible or ethically problematic. It passed the top fifty to the next team, the SimulationAgents. These are the lab rats. They built complex digital models of the tumor cells and began running virtual experiments, testing what would happen if you introduced a compound to block this protein or enhance that one. They ran millions of simulated trials, charting efficacy, toxicity, and potential side effects.

Finally, a single ReportAgent gathered all the findings. It summarized the entire process, wrote up a detailed analysis of the three most promising drug candidates, included interactive 3D models of the molecular interactions, and even drafted the introduction for the academic paper Evelyn would eventually write.

The whole process took four hours.

It would have taken 8-months before.

Evelyn’s job hasn’t been eliminated. It has been elevated. She is no longer a laborer in the salt mines of data. She is the expedition leader, the one who looks at the map and says, “Let’s see what’s over that mountain.”

The swarm does the climbing. That’s the key.

This same pattern is exploding everywhere. I know a young filmmaker in Los Angeles named Leo. He has incredible ideas but the budget of a college student. He wanted to make a short animated film about a lonely robot who finds the last living plant in a ruined city.

A decade ago, that would have required a small studio and a million dollars.

Last month, he did it with his laptop in an afternoon. Edits and everything.

One of the consequences of AI: generalists with an impassioned world view have superpowers. I’ve written an entire article on that idea:

He described his vision to his Orchestrator, which he calls “The Producer.” He specified the art style (”a mix of Studio Ghibli’s warmth and Blade Runner’s grit”), the emotional tone, and the basic plot points. The Producer then assembled his digital studio.

A ScriptAgent generated three versions of the story. A StoryboardAgent turned the chosen script into a sequence of comic-book-style panels. CharacterAgents and EnvironmentAgents designed the sad-eyed robot and the beautiful, crumbling cityscapes. AnimationAgents rigged the models and brought them to life, while AudioAgents composed a heartbreakingly perfect piano score and added the sound of wind whistling through broken buildings.

Leo wasn’t a passive observer.

He was the director.

He was in a constant dialogue with his team, tweaking, refining, and curating. “Make the robot’s eyes a little bigger.” “Can we have more dust motes floating in this sunbeam?” “That third chord in the melody feels too hopeful, let’s make it minor.” He was providing the taste, the judgment, the soul.

He had the creative power of Pixar at his fingertips, managed by one person.

The most profound impact might be in the world of business.

Starting a company is an exercise in brutal friction.

You need market research, a business plan, branding, a website, a supply chain. It’s a mountain of tedious work that stops most great ideas from ever getting started.

A young woman I advise, Maria, had an idea for a company that makes sustainable, carbon-neutral dog toys BUT she is too busy to start another venture. She needs to focus.. but she’d like to execute on this idea, and she sees an opportunity to use AI.

She became a Centaur Swarm VC.

Her “CEO Agent” took her core mission and got to work. A MarketAnalysisAgent surveyed the entire pet supply industry, identified her target demographic (eco-conscious millennials with disposable income and a beloved golden retriever), and suggested a price point. A SupplyChainAgent located three vetted suppliers of recycled, non-toxic materials in Southeast Asia. A BrandingAgent generated a company name (”TerraPaws”), a logo, and a full brand identity kit. A WebDevAgent built a gorgeous, fully functional e-commerce store, and a MarketingAgent wrote the first month of social media content.

The swarm built her entire company in a weekend. Now, she can see if the company has GTM validation before committing more resources.

If the market proves itself, Maria’s job now is to be the founder. To manage the brand’s story, to ensure product quality, and to dream up what TerraPaws does next.

This is the world we are now in. This is what I invest in.

It requires a total reimagining of what we consider to be “work.” For a century, human value has been tied to the execution of tasks. Your skill was your ability to do a thing, whether that thing was writing code, analyzing a spreadsheet, or designing a bridge.

That era is over.

The new master skill is not execution. It is direction. Your value is no longer in your hands, but in your head. It is measured by the quality of the questions you ask, the clarity of your goals, and your taste in curating the firehose of output the swarm can generate for you. We are all being promoted from laborers to managers. Or, more accurately, from musicians to conductors.

Of course, this isn’t a perfect utopia just yet. The challenges are immense and fascinating.

How do you coordinate the work of 40 agents without them tripping over each other? It’s a fantastically complex problem.

How do you ensure the system is secure? A single malicious agent could poison the work of the entire swarm.

And the biggest question of all is one of explainability. If a swarm of agents produces a truly revolutionary scientific discovery or a billion-dollar business strategy, can we trace its logic? Can we understand why it made the choices it did? Or do we have to simply trust the output of a black box that thinks in ways we can no longer follow?

These are the great puzzles of our time. But they are solvable.

I often look out my office window, down at the bustle of the city below. All those people, hurrying to their jobs, performing their specialized tasks. An ecosystem of collaboration. A human swarm. For the first time, we have built a tool that mirrors that complexity, that mirrors the very structure of our own intelligence.

The Centaur, that lonely hybrid, has been joined by an army.

Friends: in addition to the 17% discount for becoming annual paid members, we are excited to announce an additional 10% discount when paying with Bitcoin. Reach out to me, these discounts stack on top of each other!

Thank you for helping us accelerate Life in the Singularity by sharing.

I started Life in the Singularity in May 2023 to track all the accelerating changes in AI/ML, robotics, quantum computing and the rest of the technologies accelerating humanity forward into the future. I’m an investor in over a dozen technology companies and I needed a canvas to unfold and examine all the acceleration and breakthroughs across science and technology.

Our brilliant audience includes engineers and executives, incredible technologists, tons of investors, Fortune-500 board members and thousands of people who want to use technology to maximize the utility in their lives.

To help us continue our growth, would you please engage with this post and share us far and wide?! 🙏