Intelligence Was Never One Thing: The Paradox That Shaped How We Built Vantage

The most useful idea for deciding what machines should do in your firm is almost forty years old. It is the cornerstone of why we built Vantage, and most of the industry is reading it backwards.

The hardest problems in intelligence turned out to be the ones a toddler solves before breakfast.

In May 1997, a computer beat the greatest chess player alive. Chess was supposed to be the summit of human intellect, and Deep Blue climbed it. Yet no machine on Earth that year could have walked into the room, recognized the board, and moved a pawn without knocking over the clock. Nearly three decades later the pattern holds. AI passes the bar exam and still cannot reliably fold a stranger's laundry. This inversion has a name: Moravec's paradox. It is the single most useful map anyone has for deciding where AI belongs in a firm, a workstream, or a life, and it is the reason Vantage exists in the shape it does.

The hard problems are easy and the easy problems are hard

In 1988, the roboticist Hans Moravec observed that getting computers to perform at adult level on intelligence tests or board games was comparatively easy, while giving them the perception and mobility of a one-year-old was somewhere between difficult and impossible. Picking up a penny off a table is still a computationally intense task for an AI-powered humanoid robot, yet a toddler can do this with relative ease. Marvin Minsky made the companion observation: we are least aware of what our minds do best. Steven Pinker later compressed the whole field's experience into one line: "the hard problems are easy and the easy problems are hard."

In plain English: the things that feel difficult to us, calculus, chess, drafting a benchmark analysis, feel difficult precisely because they are evolutionarily new. We perform them with slow, conscious, effortful attention. The things that feel effortless, recognizing a face, reading a room, catching a falling glass, feel effortless because they run on machinery evolution has been polishing for half a billion years. Perceived difficulty and complexity of a task is a terrible guide to actual computational difficulty and complexity. It is almost perfectly inverted.

Evolution rigged the game

Descend one layer and the paradox stops being a curiosity and becomes an engineering law. The difficulty of automating a human skill tracks the evolutionary age of that skill, not how impressive it looks on a resume. Perception and motor control have been under continuous optimization since the first vertebrates, on the order of 500 million years. Language is perhaps 100,000 years old. Writing is about 5,000. Formal logic is roughly 2,500.

Put that on a calendar. If the development of these capabilities were compressed into a single year, vision and movement start on January 1st. Language shows up around 10 pm on December 31st. Writing arrives five minutes before midnight. Formal reasoning, the crown jewel of white-collar work, occupies the final two and a half minutes. Estimates on the exact timelines vary. The asymmetry does not.

If we zoom in on the final minutes, they are not uniform. One of the hardest tasks for a firm is fusing dozens of weak, noisy, correlated signals into a single calibrated probability. Weighing burn rate against cohort retention against churn against market timing against a thousand faint echoes of deals that came before, and outputting one number you would stake capital on. Firms have tried to staff analysts to solve for this, but mechanically it’s not the right fit. This is Bayesian inference at scale, and it did not exist as a formal skill before the eighteenth century. On the calendar, it is the last couple of seconds before midnight and it’s why automation should start precisely where we are weakest.

This is also why the skills evolution perfected are intuitive, unconscious, and massively parallel, while the skills we acquired last are taxing, conscious and serial. And here is the twist that matters for anyone building a firm: the recent skills are the easy ones to replicate precisely because we can articulate them. We can write down the rules of chess and the details of the share classes of a cap table. Nobody can write down how a great investor reads a founder across a dinner table, because that skill predates writing by a hundred thousand years.

Perceived difficulty is a terrible guide to computational difficulty. It is almost perfectly inverted.

The second act nobody expected

For thirty years the paradox was a robotics story. Then large language models gave it a second act, and four things changed at once.

Language fell. Reading, writing, summarizing, and pattern-matching across text moved decisively to the machine side of the ledger. Fluent prose is now abundant. As we have written before, fluency is not judgment, but fluency at scale is real and it is new.

Context stopped being scarce. Human working memory holds roughly four items at once, a limit cognitive scientists have measured for decades. A modern model's context window holds on the order of a million tokens, roughly 3,000 pages, not skimmed and summarized but held simultaneously, every clause available at all times. That is not a better analyst's memory. It is a different kind of memory.

Attention became free. A machine can run scheduled jobs against every news source, every filing, every secondary signal, every portfolio company, every morning, without fatigue and without a Friday afternoon. Human attention is the scarcest asset in any firm. Machine attention is now effectively unlimited.

The temptation arrived. Because machines now speak fluently, it became easy to believe they can now do everything, and an entire generation of AI software is being built on exactly that belief.

"Replace the humans" is a category error

Variations of the full-replacement pitch are being evaluated in almost every industry. Payroll is the largest line item in most professional firms. The demos are dazzling. And some tasks genuinely do automate end to end, cleanly and completely. If intelligence were a single axis, a dial that goes from mouse to human to superhuman, then replacement would just be a matter of waiting for the dial to turn.

But Moravec's entire point is that intelligence is not a single axis. It is a landscape with radically different terrain, and machines and humans hold high ground in different regions. Build as if the dial exists and the mistake gets poured into the foundation. No model upgrade fixes a foundation.

Approach	How it works	Where it breaks
Full replacement	Automate the entire function and remove the humans from the loop	Puts AI in the judgment seat, the one region where it holds no advantage, and removes humans from exactly the seats where they do
Autonomous decision engines	Model output executes directly: the AI trades, allocates, decides	Produces conviction without an owner. When it is wrong, and it will be, nobody can say why and nobody can answer for it
One synthesized assistant	A single model reads everything and returns one smooth, blended answer	Averages away disagreement. The most valuable signal in any real team is where the specialists diverge, and synthesis erases it

Each of these draws the boundary between human and machine according to economics or enthusiasm. None of them draws it according to where the actual advantages sit.

The Moravec allocation

There is a better rule, and it fits in one sentence: give each kind of intelligence the ground where its advantage is asymmetric, and treat the boundary between them as the most important design decision in the system.

On the machine side of the boundary, the advantages are not marginal. They are categorical. Holding an entire data room, a decade of portfolio history, and every prior memo in working memory at once. Watching every relevant source on a schedule that never slips. Recognizing patterns across corpora no human career is long enough to read. Benchmarking against reference sets that update continuously. Maintaining a competitive map that evolves every single day instead of every quarter. These are not jobs machines do somewhat better. They are jobs the human cognitive architecture and hardware was never built to do at all.

On the human side, the advantages are just as categorical, because they tap into age-old aspects unique to the human experience. Judgment under irreducible uncertainty, where the data ends and the decision still has to be made. Trust built across a table, and the thousand unconscious reads that inform it. Intuition that this Founder will run over obstacles and through walls, over and over again. Taste, including the contrarian conviction to act against the pattern the data suggests. Relationships that compound over decades. And accountability, which is not a skill at all but a property only a person can have: a decision needs an owner, and a model cannot be one.

Notice what this rule refuses to do. It refuses to treat the human as a fallback for whatever the machine cannot do yet, and it refuses to treat the machine as a junior employee. Both framings are dial thinking. The Moravec allocation is terrain thinking.

How we built Vantage around the paradox

Vantage is what happens when you take this allocation literally and build an investment firm's operating system on top of it.

Start with what we will not do. Vantage does not make investment decisions. It is not an AI trading platform, and it never will be, not because the technology is immature but because the architecture is wrong. Decisions live on the human side of the boundary, where accountability lives.

On the machine side, we did not build one assistant. We built more than thirty individual agents, each a specialist in its own discipline: conviction analysis, benchmarking, valuation, comparable events, memo work, and the rest of the disciplines a firm actually runs on. Critically, these agents disagree with each other, and the platform surfaces that disagreement instead of blending it into a smooth consensus. A real team's value lives in its arguments. We refused to build software that averages the arguments away. The result is the best team we could possibly imagine placing around yours: tireless, specialized, comprehensively read, and pointed at exactly the terrain where machine intelligence dominates, so that your partners can spend their finite attention on the terrain where human intelligence does.

And we hold ourselves to the same allocation. Vantage itself was built with AI coding agents, a true collaboration run by the same rule we sell: the agents did the categorical heavy lifting of writing the code, and our team reviewed every surface that shipped, asking the same question each time. Is this the most efficient and intuitive way to present this information to a human? The machines supplied the velocity. The humans held the taste. We think that is simply what building tools looks like from now on.

Many shapes of intelligence

The deepest implication of Moravec's paradox is not about software. It is that intelligence was never one thing. A toddler's glance, a grandmaster's calculation, a founder's instinct, and a million-token context window are different shapes of intelligence, each extraordinary in its own terrain and unremarkable outside it. For seventy years we ranked them on a single scale, and the paradox is the evidence that the scale was always wrong.

So the interesting question of the next decade is not how much human work machines can absorb. It is whether we have the discipline to notice which shape of intelligence each problem actually calls for, and the wisdom to put every gift, human and machine, on the ground where it is strongest.

Firms that manage this well will not win by adding headcount. They will win on what sits beneath their people, the infrastructure that decides what each partner ever has to carry in their own head. That is what we built Vantage to be, and it arrives in two halves. One is the operating system: the fleet of specialists that holds the machine side of the boundary and never sleeps. The other is the operating team: a bench of expert operators your partners tag in as the work demands. Both run underneath your firm, so the people you already trust are freed for the terrain that was always theirs: the judgment, the relationships, the reads no model can make. The machines take the coverage. Your team keeps the conviction.

That is the future we are building toward. It starts with a forty-year-old idea and a toddler who can do something no machine can.

Vantage puts AI where it holds the asymmetric advantage and keeps humans where they are irreplaceable. The judgment and the decision stay yours.

Conviction Made Citable.