Prediction Markets - The Truth Seeking API

Welcome to Fintech Brainfood, the weekly deep dive into Fintech news, events, and analysis. You can subscribe by hitting the button below, and you can get in touch by hitting reply to the email (or subscribing then replying)

Weekly Rant 📣

Prediction Markets Are Here To Stay. How We Build Them Matters.

If you want to see what every financial app will look like in a few years, look at Robinhood today.

They offer predictions on sports, crypto, and equities in a single dashboard. 

Are we democratizing gambling addiction, or have we shifted entirely to a new model of markets?

Corollary: 19th-century America was both fascinated and appalled by futures contract trading, with many considering it tantamount to gambling. Yet today, it is widely considered to be a cornerstone of our global trade. Prediction, too, is off to a controversial start.

Like all things in life, many things can be true at once. Prediction markets and events contracts are here to stay, and will come to define the future of markets and investment. But they’re also at the epicenter of a troubling trend towards consumer gambling addiction and harm. 

Yet the value of the information, and to the consumer of contributing information, has never been more needed. This information utility is the durable value of prediction markets. This information is also accurate.

In the post-truth, AI-generated slop era, truth-seeking markets tell us things the AI can’t. And over time, that will come to define the next era of finance and society.

Prediction markets are already big businesses. 

Analysts estimate Robinhood's revenues from its event contracts are annualizing at over $200 million in September. 

Kalshi and Polymarket recorded a $1.4 billion trading month in September of 2025. Both companies also announced record funding as Polymarket raised $2 billion from the NYSE Intercontinental Exchange (ICE), at a $9 billion valuation, with Kalshi announcing a $300 million at a $5 billion.

This product is becoming mainstream. Per The Block

Events contracts let you buy or sell the probability of something happening by a deadline, like sports, politics or even the weather.

Polymarket became particularly famous during the 2024 Presidential election for correctly calling the outcome of the election. Since then, these charts have entered mainstream consciousness, used by everything from business news TV shows to sports coverage.

Contracts range from predicting the likelihood of ceasefires to whether Jerome Powell will cough during a speech.

As product launches go event contracts have become wildly popular.

Why have prediction markets suddenly become popular?

In an increasingly geopolitical, uncertain world, where divided media and truth are less trusted than ever, markets, as cynical and profit-seeking as they may be, offer a form of quantifiable truth that very few other sources can.

Depending on who the user is, their reason for using prediction markets may run a spectrum of pure fun/gambling to using the data to help drive corporate and government decision making:

• Professional investors can adjust their market exposure: Election outcomes and geopolitics can impact commodity and stock prices. Also the wider market view on prices is a useful proxy for market sentiment. (e.g. Kalshi estimates a 92% probability of a 25bps rate cut in October)

• The data itself is also useful broadly: For media and investors who have complex stories to tell / models to update. A feed from prediction markets gives a new data element that is proving durable.

• Consumer speculation drives much of the value for others: But when you can bet on if someone will cough, its a fine line between rewarding pro-sumers for their research abilities and simply repackaging sports betting. When Kalshi defines “sports” as an event contract, suddenly, sports betting sits right inside your finance app.

The benefits for professional investors and media are a large driver for why this asset class is growing, but the question is whether this is at the expense of, or a benefit for, consumers.

The consumer rationale is much broader than “YOLO”

It’s overly simplistic to say “because gambling is addictive,” even if there is some truth to that, it massively undersells the breadth of what’s happening. 

Consumer speculation has gone mainstream, and the line between gambling and investing is blurring. 

• In 2018, the U.S. Supreme Court opened the door for states to legalize sports betting. 

• In 2021, the memestock mania overtook consumers during the pandemic, with stocks like $GME (Gamestop) becoming wildly popular.

• And not to forget, crypto itself is subject to massive volatility, with peaks in consumer activity when the price is higher.

You can barely watch any sports without seeing an ad for how to speculate on the outcome (it’s the same in the UK too FWIW).  Now, not all speculation is gambling, but the line is blurry.

Consumers are becoming financially active much younger.

Generation Z grew up in the shadow of the financial crisis. While millennials could make it into the property market if they were incredibly successful, Gen Z enters the workforce at the tail end of an asset price bubble that has run for decades. Homes have never been so far out of reach. These two generations also grew up with video games, where XP, rewards, and “loot” mechanics are a part of the common psyche. 

If you’ve played Minecraft, Roblox, or any MMORPG, suddenly resources, pricing, and markets are not such an alien concept. They’re background.

The social contract for Gen X and Boomers worked. Get a property, get a good job and you can probably retire comfortably. The social contract for the younger generations is one with declining prospects, indebted nation-states, and where property is probably a bad investment overall compared to sitting in the S&P 500 index and compounding.

The rise of finance as content and fintech apps like Robinhood.

The memestock craze was typified by “Roaring Kitty,” AKA, Keith Gill an investor, educator, and content creator who shared his insights on stocks on Reddit and YouTube. The platforms became an opportunity to share analysis on why a stock would move and engage in discussions with users. 

Robinhood sits at the crossroads of many of these trends, allowing consumers to invest with lower friction and cost than ever before. During the pandemic, these stocks became mainstream news and even entertaining. The speculation of price action became a social phenomenon. 

This is something all the new speculative assets share. They all blend, finance, hope, entertainment, and a consumer base for whom the traditional 60/40 portfolio and social contract are no longer a path to success. When you add it up, you have a consumer base that’s less likely to have massive student debts and faces high asset prices.

The Value: Truth That Costs Something

The internet became noisy. Truth became claims. And consumers have lost trust in media.

Prediction markets force participants to put money behind their beliefs. That pressure creates a much-needed alternative to the loud social media or TV pundit world. When you can lose money for being wrong, you think differently about what you claim to know.

This is why they matter beyond their dollar volumes. Social media lets anyone claim anything. TV pundits face no consequences for being wrong. But prediction markets require commitment.

Prediction markets are a structurally different business model from sports betting.

• Sportsbooks set odds to ensure they profit and limit winning players. When customers win, sportsbooks lose.

• Prediction markets allow the market to set the odds, with no upper limits, they win if there’s more activity on the platform.

The old model needed millions of users to slowly lose. Sports betting sites will slash their revenue guidance if their players win. Prediction markets are more incentive-aligned, and they create value through data.

The real economy can already use this data:

• A farmer deciding crop allocation looks at grain-price futures and prediction markets on tariffs or weather policy

• Companies use internal or external markets to gauge product launch demand, regulatory risk, or adoption rates

• Hedge funds are using prediction markets for macro positioning

For banks, fintech companies or anyone managing risk, ask yourself:

• If you're in insurance or underwriting, can you feed your models with event-priced signals?

• If you're in investments, how can the events data (not contracts) inform your customers' decision-making?

• For regulatory teams, how can understanding timelines on future rulemaking help you prioritize?

Matthew Prince of Cloudflare put it perfectly, 

In a world where AI can generate infinite plausible stories, what compounds is decision quality. Consumers get a single-number probability that's often more accurate than media narratives or polling. The true benefit is reducing uncertainty in decision-making, not just making money off bets.

When Goldman starts reporting prediction market data in research notes, corporate boards start asking, "Why aren't we using this?" When Fortune 500 companies standardize on it, governments can't ignore it.

A cynic might say this is just repackaged gambling. But that misses the point. Just as 19th-century America was fascinated and appalled by futures contracts, prediction markets create an information utility that's too valuable to ignore.

The point isn't that everyone should gamble. It's that we finally have a way to price belief that forces skin in the game. In an era of costless claims, that matters.

This progression is inevitable. But to get there, we should predict the negative consequences and mitigate them now.

The Risks Are Real, But We Have Agency

Whenever we do something new in the world, we unleash unintended consequences or negative externalities. What’s good for profits might not be good for society or the economy. 

Consumer protection concerns are real and well-documented. The gambling addiction data shows a 61% increase in help-seeking after sportsbook legalization. One neobank exec told me their #1 non-discretionary spend category is now sports betting and prediction markets. And a key driver for prediction market usage is sports.

This keeps me up at night. Who's incentivized to protect vulnerable consumers besides the government? The answer can't be "restrict access" because that creates a two-tier truth-making system and the very kind of lockout that created the memestock madness in the first place. But it also can't be "unrestricted access" because the harm is measurable.

The solution has to be infrastructure-level design:

• Cool-down periods and position limits based on verified income

• Affordability checks before high-stakes participation

• Rewarding prediction quality over bet size

Market integrity questions are trickier. Insider trading rules don't clearly apply to "Will Jerome Powell cough?" But they should probably apply to "Will FDA approve this drug?" The line needs drawing, but it's not obvious where. In a world where we take liberties with financial data, there’s a Nancy Pelosi portfolio index. The term “market integrity” risks becoming performative.

Then there's the systemic question: Kalshi offers 3.75% APY on idle cash. That's a deposit flight from traditional banks. Is that a bug or a feature? Depends on whether you think prediction markets are infrastructure or products, and what the center of gravity is for consumers in a decade. I believe (predict) it inevitably moves towards everything wallets.

Lastly, there’s the regulation question. Kalshi has a CFTC-regulated Designated Contract Market (DCM) license, Polymarket acquired QCX LLC in July to achieve the same license, and the CFTC issued a "No-Action Letter" in September. Both companies use the self-certification mechanism to list new contracts without prior approval as long as the CFTC does not object. On the surface, this is “solved,” but in reality the consumer harms could create a future backlash

These are valid concerns, but they’re also addressable. Sometimes the voices against prediction markets and pro-consumer health are ones whose solution is usually “this is bad, it should be limited or banned.” And I think that’s unrealistic. There’s too much value for institutions in the data, too much demand from consumers. Our job, therefore, is to build things the right way.

Predicting and Preventing the Negative Consequences.

If prediction markets become infrastructure for establishing truth, who gets to participate?

The traditional answer is "protect consumers by restricting access." But that creates a two-tier system: sophisticated actors who can participate in truth-making, and everyone else who can't.

This matters because participation in prediction markets isn’t just gambling—it's participating in the formation of consensus reality. Excluding people from that is a new form of disenfranchisement. Which, ironically, is the exact problem that caused most of society to turn to high-risk investment in the first place. The sense that the game was rigged, and they’re locked out.

But unrestricted access creates genuine harm. The gambling addiction data is real. The deposit flight from banks is real. The potential for manipulation is real.

We need infrastructure for responsible participation, not gatekeeping.

The wallet is the UI for prediction markets, but we have the opportunity to build friction like limits, cool-off periods, and self-exclusion into the infrastructure from day one. And friction can be rewarding.

(Games have figured this out. 5 loot boxes per day. Daily tasks. Go look at the battle pass of any video game to figure out how having people get less feels like more. Friction is the product. Anticipation can become a feature, not a bug.)

The line between investor and citizen has blurred. Participation in markets is participation in society's consensus-formation process.

Is this utopian or dystopian?

Both, potentially.

The difference between these futures depends on the choices we make now.

• Do we build consumer protection into the infrastructure, or bolt it on later after the harm is done?

• Do we treat prediction markets as infrastructure for collective decision-making, or as just another product to monetize?

• Do we ensure broad participation, or create a two-tier system where only the wealthy influence consensus reality?

My Prediction: We will get this right.

For the last few decades, we've been living through the breakdown of our truth-making mechanisms. Social media showed us that anybody can have a platform. And now AI is about to show us that anybody can generate an infinite number of plausible stories.

Prediction markets won't solve this completely. But they're the best mechanism we've invented yet for creating enough consensus to act.

They force commitment. They reward accuracy. They create Darwinian selection for truth-telling.

Prediction markets are here. The question isn't whether they become infrastructure for truth-making, but whether we build them with protections baked in or bolted on.

The difference determines whether we get a more democratic mechanism for collective decision-making, or whether we financialize every aspect of human existence while excluding those without capital.

We have maybe 18-24 months to get the infrastructure right before the backlash forces heavy-handed regulation that breaks what's valuable about these markets.

After that, we'll be stuck with whatever we built.

Build accordingly.

ST.

4 Fintech Companies 💸

1. Cycles - The Open Clearing Protocol 

Cycles provides a clearing mechanism for onchain finance. Instead of requiring traders or borrowers to over-collateralize in order to lend, it clears obligations with its algorithm. Its algorithm is a privacy-preserving map (graph) of obligations onchain. If Alice owes Bob $10, and Bob owes Clive $10, and Clive owes Alice $10, Cycle can clear those obligations without requiring collateral. It calculates the possible closed loops and nettable positions (with TEE and ZK Proofs), then sends parties a daily journal and sets off notices, meaning only net balances are outstanding.

🧠 Clearing is a critical infrastructure for payments that works by centralizing risk. Cycles has a very different approach. Clearing banks historically help many other banks and non-banks complete a payment by making sure every payment is going to the right place. They solve the many payers, many payees problem, by taking on the risk in the middle. Smaller banks and non-banks park collateral with a clearing bank, which will directly access the central bank or correspondent banking. Cycles doesn’t take on the risk centrally. Instead, it “respects the graph.” It uses the blockchain to figure out the nature of debts and obligations.

2. Cryptomesh - The Staking Gateway

Cryptomesh allows users to “stake” their crypto assets across multiple chains through a single gateway. This avoids the pain many institutions have of dealing directly with chain-specific staking interfaces and services. It can also help move assets between chains to maximize reward opportunities.

🧠Staking is becoming an asset class. Staking is how networks like Ethereum and Solana secure their network and create new token distributions. Consistently, these tokens deliver between 3% and 4% APY. They’re considered safe and largely uncontraversial by large institutional buyers, but they’re complex to use. Staking rewards are available in many crypto apps, but over time are becoming an asset class investors want to trade in all forms of marketplace. I believe in 3 years, staking will be as common as crypto/buy/sell hold is today.

3. Falcon Finance - Turning anything into collateral

Falcon lets users deposit stablecoins, crypto assets or tokenized real-world assets to receive USDf. This receipt “stablecoin” acts as yield-bearing collateral in onchain markets. This means users don’t have to sell their underlying access to access collateral.

🧠Collateral coins are a subcategory of stablecoins. Yield-bearing collateral is uncontroversial. In financial markets its common to use Treasuries or other low-risk assets as collateral to trading activity.  USDf differentiates with a broad approach to what can be collateralized. 

4. Clove - AI-Powered Wealth Management for Everyone Else (UK)

Clove aims to provide financial advice to the mass market with a combination of AI and human support. The goal is to make financial planning affordable and effective for growing families, entrepreneurs, and young professionals.

🧠Financial planning sucks, I hope Clove fix it. It's often an exercise in giving someone a lot of personal data and documents to hear some regurgitated best practices you could have found on Google. Most financial advisors then ultimately cluster clients into a handful of funds. The paperwork chasing and analysis can be streamlined with AI, and humans provide the empathy wrapper we sometimes need to really listen and do what’s in our best interests.

Things to know 👀

1. Erebor gets Preliminary OCC Approval. Bridge Applies for a Trust Charter.

The stablecoin-native bank backed by Peter Thiel and Palmer Lucky, Erebor, is the first to receive preliminary, conditional approval from the OCC this administration. Bridge aims to use its Trust charter for custody, issuance, and direct management of stablecoin reserves.

🧠 Can Erebor fill the gap left by SVB? When SVB fell, there was no longer a large, tech-focused, entrepreneur-focused brand. This is exactly what Erebor is targeting.

🧠 Erebor will be a test case for how regulators treat the gap between digital assets and TradFi. Expect the big banks to watch closely.

🧠 As Alex pointed out, it’s most likley the Trust Charter isn’t for custody because the separation of concerns restrict self dealing. (money moving vs being a custodian)

🧠 Therefore, the Trust Charters are for Fed Master accounts. Which let you settle as an on/off ramp instantly. This improves unit economics and makes the on/off ramp more seamless.

2. Brex is now Embedded in the Oracle’s Fusion Cloud ERP

Brex Cards will be built into “Fusion Payables” the AP system used by some of the worlds’s largest corporations. So any supplier payment now comes with all of the card control and spend management features (like merchant locks).

🧠 Brex Embedded is smart GTM for enterprise clients. While Ramp is building an AP system, Brex is embedding in everyone elses.

🧠 I know of a bank who tried to do something similar 5 years ago. But they were forcing every single client to open a new account before they could make a payment. Embedded finance is as much about the experience you create as who you partner with.

🧠 In the next 3 - 5 years every ERP, HR, and Travel system will become modular. Brex’s counterpositioning here is interesting. Growth companies wanted the all in one. Enterprise companies need flexibility. Choose your fighter.

3. Visa Trusted Agent Protocol (TAP Launched.

Visa just launched the Trusted Agent Protocol after what it says is a 4,700% surge in AI-assisted shopping. Visa's solution: cryptographic signatures for agents. Three pieces of data pass from agent to merchant. Agent Intent: "I'm here to buy, not scrape"Consumer Recognition: "This person has shopped here before" Payment Info: "Here's how they want to pay." Visa. Adyen, Stripe, Shopify, Coinbase, Microsoft, Worldpay are all early partners.

🧠Note the timing: Launched right before earnings, but not long after a blizzard of other protocol announcements.

🧠Note the politics: It’s designed to be compatible with Stripe/OpenAI’s “ACP” and Coinbase x402. But there’s also names missing (no Google, no Open AI). Now, you can’t announce everyone in every press release. But it’s already looking a bit “format wars” out there.

🧠Visa is in a different starting place. This is a trust protocol not merely a commerce protocol. Visa’s brand is about saying, this transaction can happen and if something goes wrong, here’s the rules we use. That’s needed no matter what.

🧠We have the protocols we need frameworks and tooling. There’s barely any scaffolding for agentic commerce. PSPs are trying to pick off various protocols but its too soon to say who wins.

🧠OpenAI could be a kingmaker. While Anthropic is compounding revenue from enterprise, of ChatGPTs 800m consumers only 5% subscribe. They either need that attach rate to grow, or commerce / ads to take off.

Good Reads 📚

1. The new commerce tax, ads or take rate?

Ad expense and payments take rate are an interchangeable form of “tax” on digital sales for merchants. Youtube “taxes” creators 45%, Apple “taxes” payments 30%, or card swipes by 15bps. In a ChatGPT world, they could start with a low tax (say 2%) for every payment they facilitate but they’re blurring the line between ad and payment take rates. With more volume they’d gain pricing power over time.

🧠ChatGPT is converting between 2x and 6x better, it’s a no brainer priced at 2%. That over time could shift volume to the platform, and as that happens, ChatGPT gets pricing power.

🧠This is the first well thought through public analysis of the unit economic impact on merchants, ads and payments I’ve seen. Really well done and worth diving into the longer substack piece here. 

Tweets of the week 🕊

That's all, folks. 👋

Remember, if you're enjoying this content, please do tell all your fintech friends to check it out and hit the subscribe button :)

Want more? I also run the Tokenized podcast and newsletter.

(1) All content and views expressed here are the authors' personal opinions and do not reflect the views of any of their employers or employees.

(2) All companies or assets mentioned by the author in which the author has a personal and/or financial interest are denoted with a *. None of the above constitutes investment advice, and you should seek independent advice before making any investment decisions.

(3) Any companies mentioned are top of mind and used for illustrative purposes only.

(4) A team of researchers has not rigorously fact-checked this. Please don't take it as gospel—strong opinions weakly held

(5) Citations may be missing, and I’ve done my best to cite, but I will always aim to update and correct the live version where possible. If I cited you and got the referencing wrong, please reach out