Bloomberg didn't win by putting prices on one screen

KosmosManifesto

In 1981, a bond trader could not answer a simple question: is this bond cheap? Not because the prices were secret. Because the comparison was impossible. You called three desks, wrote yields on a legal pad, and did the arithmetic by hand while a counterparty held the line. The trader who could do that math fastest won, and almost nobody could do it fast.

Michael Bloomberg built a machine that answered the question. In December 1982 the first one went into Merrill Lynch, who bought twenty of them and a 30% stake for $30 million. Today there are more than 325,000 seats at $31,980 a year.

He didn't win by putting prices on one screen. He won by making the terminal the place where a decision got made and defended. The data was raw material. What he sold was the confidence to act on it, in the seconds when acting mattered.

Forty years later, the raw material is free and the confidence is still scarce.

The first market for being right

Every other exchange trades a claim on something. A share of a company. A barrel of oil. A currency pair. Prediction markets trade the proposition itself: will the Fed cut in September, will the ceasefire hold, will the drug clear phase three.

The asset is a belief, and it settles. No quarter to blame. No thesis that was early. No story to tell afterward. You were right or you were not, on a date, in public.

A pundit is never repriced. These markets reprice every second, by people with money on the line.

They are the only forecast that costs something to be wrong about. For most of history the best read on what happens next lived inside an institution and was sold to whoever could afford the relationship. Now it is a public number that anyone can see and anyone can disagree with, in size.

Every contract is a derivative on a real-world risk that never had a market. A farmer can hedge rainfall. An importer can hedge a tariff decision. A campaign can hedge an outcome. A hospital can hedge a policy change. These are enormous, ordinary risks, and until now there was no instrument to lay them off. Equities let you hedge companies. Event contracts let you hedge the world.

The tooling has to catch up

The infrastructure is arriving fast and a lot of it is good. What is still missing is the layer above the data, the one that decides what any of it means.

IntelligenceBeing writtenWhich contract deserves your attention · what it pays on · why it moved · how sure to be
DataShippedUnified APIs · screeners · arb scanners · cross-venue terminals · SDKs
VenuesShippedPolymarket · Kalshi · 61,000 live contracts · whatever lists next
The stack as it stands. Venues arrived, then the data layer underneath them. The layer that turns a price into a decision is the one still being written.

Equities got Bloomberg. Crypto got a hundred terminals in eighteen months. Prediction markets are the most information-dense asset class ever built, and their intelligence layer is still in its first year. That is the gap.

When the thing you hold is your own judgment, the constraint stops being information. It becomes conviction.

Whether you believe your read enough to size it, hold it through a drawdown with no floor under it, and not fold when a headline lands at 2am. That gap, between having a view and having a position, is what we are building for.

What we are building

Not a screen. An intelligence.

  • Personalized agents. Autonomous research and execution while you sleep, inside rules you set.
  • Quant research. Base rates, calibration and how contracts like this one have resolved before. The work behind your read, not a call to follow.
  • Trader intelligence. Who is right luck-adjusted, when they enter, how far they move the market, and whether they are copyable at your size.
  • The live tape. News tied to the exact contract it moves, in seconds. Smart-money orderflow as it prints.
  • Microstructure. The book, not the last print. Depth, spread, and what your order actually costs to fill.
  • Every exchange. Polymarket, Kalshi, and whatever lists next.

The point

Markets have paid people for being right about companies for four hundred years. Prediction markets pay people for being right about the world. That is a bigger idea, and we are at the very beginning of it.

The people who will be great at this will not be the ones with the most screens. They will be the ones who can look at a number everyone agrees on, know something everyone does not, and put real weight behind it. That has always been what separates a trader from a spectator. It has never had proper tools.

Bloomberg gave a generation the confidence to price a bond. We are building the thing that gives this one the confidence to price what happens next.

More from Kosmos