Someone just revealed who makes Polymarket prices accurate. Every Polymarket contract ends at $1 or $0. The price before resolution is supposed to reflect the market’s live estimate of the outcome. A 2026 study on prediction market accuracy by Roberto Gómez-Cram, Yunhan Guo, Theis Ingerslev Jensen, and Howard Kung analyzed all of Polymarket's transaction data. They tested whether Polymarket prices are shaped by crowd wisdom, meaning many traders each add a small piece of information, or by a small group of traders who move prices closer to the truth before everyone else reacts. To separate skill from luck, they used a sign randomization test. They kept each trader’s real markets, timing, size, and prices, then randomized the direction of the trades. The finding: only around 3% of traders are skilled. Using the paper’s method, you can identify skilled traders, track their activity, and turn it into a signal. Polymarket’s public API exposes user trades, prices, sizes, timestamps, and current positions, so this can be followed directly. What if the real alpha was copying informed traders?

Your risk model might be quietly costing you more money than you think. Not because you made bad calls. Because the scenarios your risk model runs to protect your portfolio are partly fiction. There is only 1 dataset of the S&P 500. 6,064 trading days. That's it. Fake correlations When generative AI produces thousands of future scenarios from that single history, it fills the gaps. With invented correlations and volatility spikes that look plausible but are mathematically wrong. The model has no idea what it doesn't know, so it makes it up...confidently. That's where the money goes. Maximum Entropy Maximum Entropy was written down in 1957 by a physicist named Jaynes. Around one idea: don't invent what you don't know. Commit only to what the data tells you. Stay uncertain about everything else. The catch: it was so slow and unusable at real scale. Days of compute per run. So the industry moved on and accepted hallucination as the cost of speed. Maximum Guided Diffusion In February 2026, researchers at ENS Paris, NYU Courant, and Capital Fund Management cracked it. They built Moment Guided Diffusion “MGD”. MGD runs at modern AI speed. But at every single step of generation, it enforces one hard rule: match the real data's statistical fingerprint. Nothing invented beyond that. No fake correlations. No invented structure. They tested MGD on 24 years of S&P 500 returns, turbulence simulations, and dark matter maps. The fat tails, the volatility bursts, the long-range dependencies all reproduced. Nothing added. MGD is what honest scenario generation looks like. This matters for traders who think their risk tools are “good enough” in violent markets.

Most founders we talk to have the same story. Months on the whitepaper. The tokenomics. The raise. They found a market maker, signed the mandate, and felt like the hard part was done. Then trading started. Volume looked fine. The MM was technically doing its job, bots were live, and quotes were showing up. But something felt off. A $50k buy was moving price more than it should. Spreads on certain venues were twice as wide as others. They asked their MM what was happening. They got a monthly summary. Here's the problem: most mandates are built to run autonomously. The MM deploys, manages, reports back. And in that model, the token team is expected to trust the process. So when something breaks, they're always one step behind. By the time they understand the cause, the damage is done. That's not a bad MM problem. That's a design problem. A token team should see what's happening in their market and understand their MM response as it happens, not 30 days later. That's what we built Chainswatch around.

Try @geoprotocol

SPEAKER REVEAL 🎙️ Ever wondered what really goes into getting your token listed? Kevin Primicerio, founder and CEO of @PerplexFi, will walk you through the fundamentals in: “Market Making 101” A must-attend for builders looking to launch. Secure your seat: arweavedayasia.net/#speakers

Speaker: Kevin Primicerio, Founder and CEO of @PerplexFi Topic: Listing your token: Market Making 101

1/ Are we pricing AI like a trend when it behaves like a productivity shock? First, capacity strain. Then a re-pricing across equities, rates, FX, and capital flows. A thread👇

Analyzing @ar_io_network protocol revenues and rewards. Since the mainnet launch 7 months ago, the network earned ~6.2M $ARIO (~$85k) and distributed ~11.5M $ARIO (~$196k) to gateway operators. Emissions are boosted still at 0.1% daily, reducing linearly to 0.05% after the first year. Early days yet permanence is already paying. Imagine these numbers when the world realizes the cloud doesn’t have to expire.

✅ Stop Loss & Take Profit are now complete! Easily set automatic exit points for your trades on @aothecomputer & @arweaveeco Take Profit: Secure your gains automatically. Stop Loss: Limit your losses, even when you're offline. Beta testers loved it, and we can't wait for YOU to try it soon! The best is yet to come.

🫡 Thanks for the salute. Arweave's tamper-proof storage is invaluable for preserving unalterable truths in a world of fleeting data.

About a year ago, after spending too much time trying to debug our code, we started writing the v0.1 of aolite. Since then we’ve spared countless of dev hours and headaches! I would encourage anyone taking part in the @aoTheComputer hackathon to try it out! It will give you an edge :)

Today's Arweave Highlights: ✦ @kprimice introduces AO Lite ✦ @ankushKun_ shares social login tool ✦ @StarGridBattle lets you play against AI ✦ @aoTheComputer hackathon announced Plus: @HorizonXBT's $ARIO microgrant guides small businesses through the permaweb

Get the alpha on AO Lite👀 Speed up testing and development with this open-source toolkit for fast, local simulation of AI trading bots. Originally built for @PerplexFi, now open source for all builders. ▶️ Watch @kprimice demo what’s next for trading automation.