What happens when a Layer 1 built for real-world assets walks into a research lab studying whether crowds can price the future better than analysts?
Pharos Network is about to find out, and the answer could decide whether prediction markets stay a sports book with a crypto wrapper or become the pricing layer for everything from corporate earnings to tokenized treasuries.
Behind the Deal
Pharos Network has formalized a research collaboration with the HKU-SCF FinTech Academy, structured inside the Master’s Capstone Project framework at HKU Business School. Eight master’s students, supervised by Dr. You Yang, Assistant Professor of Finance at HKU, will spend three months on empirical work covering the validity boundaries of collective prediction, structured modeling of event probabilities, and the application of AI models in predictive decision-making.
The unusual part is the back end. Outstanding projects can graduate into the Pharos global incubation program, which sits inside a $10 million “Native to Pharos” builder track backed by Hack VC, Draper Dragon, Lightspeed Faction, and Centrifuge. That converts a typical academic exercise into a funnel for live products on a live chain.
The HKU-SCF FinTech Academy itself is not a small partner. It was established in 2020 with HK$60 million from the Standard Chartered Hong Kong 150th Anniversary Community Foundation, hosted by HKU’s Department of Computer Science with backing from the Faculties of Law and Business and Economics.
Why Prediction Markets Are Suddenly the Story
A prediction market is a venue where people trade contracts that pay out based on real-world outcomes. If a contract trades at 62 cents, the market thinks there is a 62% chance the event happens. The price is the forecast.
The category went from curiosity to scale in 2025. Total notional trading volume crossed around $44 billion across the sector, with Polymarket and Kalshi accounting for roughly 97.5% of activity. Kalshi crossed $2.3 billion in a single week by late December. Polymarket CEO Shayne Coplan, on 60 Minutes, called prediction markets the most accurate forecasting tool humanity has built.
The interesting wrinkle is the category mix. Politics gets the headlines, but Technology and Science markets grew 1,637% year over year and Economics grew 905%, while politics grew only 43%. That shift is exactly where Pharos and HKU say they want to push the academic work.
What Actually Matters
Pharos is positioning itself as infrastructure rather than another consumer prediction app. The chain claims up to 30,000 TPS with sub-second finality through a parallel execution architecture that handles consensus, execution, storage, and data availability concurrently. That throughput matters for prediction markets because settlement happens in bursts around real-world events: an NFL game, a Fed decision, an earnings print.
The second piece is x402, an open payment standard built on the HTTP 402 status code that lets AI agents pay for services without human approval. Pharos says its X402 module is protocol-ready for agent interaction. In practice that means an autonomous research agent could pay for a data feed, settle a prediction trade, and pull a probability signal in one workflow, with no human pressing buttons. Google’s Agent Payments Protocol now includes x402 as an extension, which gives the standard real distribution.
Wish Wu, Co-founder and CEO of Pharos, framed it this way:
The essence of prediction markets lies in the accuracy of data input and value output, aligning perfectly with AI’s capabilities. Pharos aims to be not only a settlement layer for financial assets but also a verification layer for information.
Prediction market volume is concentrated in sports and politics because that is where retail attention lives, but the harder and more valuable problem is whether the same mechanism can price a tokenized treasury, a private credit pool, or a Q3 earnings number. If a Capstone team produces a working model for binary options on RWA cash flows, that is genuinely new territory.
Dr. You Yang put the academic case in plainer terms:
While the commercial potential of prediction markets is evident, trading volume remains concentrated in entertainment-driven areas like sports and political betting. This project will explore, from an academic perspective, how binary option mechanisms can be applied to RWA pricing, auction design, and corporate earnings forecasts.
The risk is the obvious one. Polymarket and Kalshi own the distribution, and CFTC scrutiny has already produced state-level pushback against sports contracts. A new chain entering this space wins by being the rails for use cases the incumbents will not touch, not by cloning them. The HKU partnership, the Ant Group engineering pedigree, and the $52 million in total funding at a $1 billion valuation give Pharos a credible shot at being that infrastructure layer, especially if the x402 module pulls real agent volume.
Final Thoughts
The most useful thing about this collaboration is that it forces a public test of an unproven claim, that prediction markets can grow up into pricing infrastructure for real assets. Academic rigor on one side, a live chain and incubator capital on the other, and three months to produce something defensible. That is a healthier feedback loop than most industry-funded research.
If even one or two of the eight Capstone projects ship into the incubator with a working model, Pharos gets exactly what it wants: proof that its chain can host a category of application that is bigger than betting and that institutions might actually use.
Don’t forget to like and share the story!