Prediction market making gets easier with frontier models and Jev
The Trueo team argues that prediction markets lack liquidity because pricing is modeling-heavy, and that connecting a few data APIs to frontier models with Jev for decisions now makes market-making strategies feasible at scale.
I can’t underscore enough how much of a tailwind AI advancements are for prediction markets. Historically futures and options have always been significantly more liquid than prediction markets because pricing them and providing delta-neutral liquidity is a lot easier. It’s super simple to run a market making strategy that just hedges exposure to an asset on any liquid venue. After all, there are 100s of liquidity sources for ETH or BTC. But it’s orders of magnitude harder to run a delta neutral strategy for the rental cost of NVDA’s chips, for example. Which is why the majority of prediction markets liquidity comes from LPs and traders that model the outcome of a specific contract, then make liquidity where they think the price should be. It’s the modelling part that used to be too intensive for even the most active and performant high frequency trading shops. But now, what once required significant engineering resources to build a custom machine learning algo from scratch is as simple as connecting a number of data APIs to a strategy that runs through the frontier models. And with new innovations like Jev, these strategies become increasingly feasible at scale. This is an area of priority interest for Trueo. I’m regularly told that if we can solve for liquidity on our platform we have an opportunity to directly competing with Poly and Kalshi. What I’m saying is that this goal has never been more achievable. And we’re going to prove it very soon.