Prediction markets are entering a new phase, and World is introducing a third approach to one of the industry’s biggest challenges: how markets decide who actually won.
Kalshi relies on a rulebook and a regulated operator. Polymarket uses an optimistic oracle where outcomes can be challenged and voted on by token holders. World takes a different route, using Chainlink data feeds to automatically settle contracts once an event reaches its defined outcome.
There is no human resolution panel, no token-holder vote and no dispute window.
That makes the process faster, but it also creates a different set of risks.
World opens its standalone prediction market platform
World opened its standalone platform at world.xyz on Sept. 9, giving access to more than one million people who had been waiting for the launch. The platform had previously operated inside the Phantom wallet during the summer.
More than 150,000 markets have already been created across sports, crypto, politics, finance, economics and culture.
The initial selection includes every NFL regular-season game, seven soccer leagues, Formula 1 races, contracts tied to the 2026 US midterm elections, and markets based on Federal Reserve policy decisions.
World also plans to expand into areas such as equities, commodities and weather.
The basic structure is familiar. Users trade binary yes and no contracts, generally priced between $0 and $1. Once the outcome is verified, the winning side settles at $1.
What makes World different is how that outcome is determined.
The platform is non-custodial, meaning it does not hold customer funds. Orders are routed to liquidity providers on Solana rather than through a traditional off-chain order book.
Settlement takes place in CASH, the dollar-backed stablecoin used within Phantom, with winning positions automatically redeemed in the user’s wallet.
Users also do not need a brokerage account or exchange registration to participate, while network fees apply when positions are opened or closed.
Three ways prediction markets can settle
The simplest way to understand World is to compare it with the two major models already operating in the market.
Kalshi: human judgment under a rulebook
Kalshi uses predefined contract rules that are applied by a regulated exchange.
A designated operator determines the outcome, with regulatory oversight and a formal complaints process behind the system.
The advantage is flexibility. If something unexpected happens, the operator can interpret the rules and potentially correct an error.
The downside is that human judgment introduces discretion. Participants may disagree with the decision.
Polymarket: an optimistic oracle
Polymarket uses an optimistic oracle model.
An outcome is first proposed, followed by a challenge period. If someone disputes the result, the matter can eventually move to a vote involving holders of the oracle’s governance token.
This removes the ability of a single operator to unilaterally decide an outcome, but it can make settlement slower and more contentious.
World: automated settlement
World removes the human decision-making stage almost entirely.
Chainlink Data Streams provide the relevant data, while the Chainlink Runtime Environment handles the settlement process once a defined event has ended or reached its deadline.
There is no resolution panel, token-holder vote or dispute period.
The result is speed and automation, but the trade-off is important: once the underlying data determines the outcome, there is no conventional mechanism for challenging it.
Automation works best when the outcome is objective
World’s model has an obvious advantage when markets rely on clean, machine-readable information.
Consider a football match. The final score is a clearly defined number.
The same applies to a Bitcoin price at a specified time or a Federal Reserve interest-rate decision.
In these cases, a dispute process can add unnecessary delays. If everyone agrees on the underlying data, an automated system can settle the contract immediately.
That can improve the user experience while also allowing the platform to support a much larger number of markets.
World has already created more than 150,000 markets, something that would be difficult to manage if every settlement required human review.
Automation also eliminates one of the biggest sources of reputational risk for prediction markets: controversial resolutions.
If the outcome is determined directly from an agreed data source, there is less room for an operator or voting group to make a judgment that participants believe contradicts reality.
But automation has clear limits
The same feature that makes automated settlement attractive can become a weakness when an event is difficult to define.
Imagine a market asking whether a ceasefire will hold.
There is no simple numerical data point that can answer the question. Someone has to determine what counts as a violation, whether a particular incident qualifies and whether a reported event is credible.
A rulebook can provide a framework for that decision. An oracle-based system can allow people to challenge it.
An automated data feed, however, cannot independently resolve an argument about the definition of the event.
What if the data feed is wrong?
There is another risk.
If an oracle reports the wrong value, reports late or uses information that does not accurately reflect what happened, an automated system can still settle the market based on that input.
Without a dispute window, there may be no opportunity to stop the settlement before funds move.
That makes the reliability of the data source one of the most important parts of the entire system.
Edge cases can also become complicated
Postponed sporting events, abandoned races, rescheduled announcements and data feeds that stop working are all situations that prediction markets eventually encounter.
Traditional rulebooks can contain provisions explaining what happens in those circumstances.
An automated platform must rely on whatever conditions were written into the contract beforehand.
And no contract can perfectly anticipate every possible real-world event.
The regulatory question remains unresolved
World’s launch also raises an important regulatory question.
The platform lists markets covering NFL games, the 2026 US midterm elections and Federal Reserve decisions — categories that are already attracting significant regulatory attention in the United States.
Kalshi operates as a designated contract market, while Polymarket has established a regulated US exchange presence. Both have faced legal and regulatory challenges despite taking formal steps toward compliance.
World’s structure is different.
It is non-custodial, does not hold customer funds, routes transactions through a public blockchain and does not require users to open a traditional brokerage account.
That could give the platform a different legal posture.
However, it does not automatically mean regulators will view the protocol as outside existing frameworks.
The bigger unanswered question is whether regulators will apply traditional rules designed for centralized exchanges to a decentralized, non-custodial system.
The launch materials also do not clearly address geographic restrictions, making that an important issue to watch as the platform expands.
Why Solana and Phantom matter
The choice of Solana is closely connected to the platform’s distribution strategy.
Solana has become a major venue for high-volume crypto trading activity, while prediction markets are increasingly being positioned as another consumer-facing application for the network.
Phantom plays an equally important role.
World initially operated inside the Phantom wallet, giving it access to an existing user base rather than requiring customers to download a new application or create a separate account.
Settlement through CASH also keeps the experience closely connected to the Phantom ecosystem.
That helps explain how World was able to build a waitlist of more than one million users.
For competitors, this is potentially more significant than the underlying settlement technology.
Kalshi and Polymarket have invested heavily in advertising, partnerships and user acquisition. World instead gained distribution through an existing wallet integration.
Chainlink could become the bigger story
World’s launch is also significant for Chainlink.
The platform uses Chainlink Data Streams and the Chainlink Runtime Environment to automate the creation, resolution and settlement process.
Another prediction market has adopted the same infrastructure for automated settlement of crypto price markets, suggesting that Chainlink could be moving toward a broader role as prediction-market infrastructure.
That matters because there is a difference between being a provider of price feeds and becoming a core settlement layer.
If multiple prediction platforms begin relying on the same infrastructure, switching providers could become increasingly difficult.
There is also a potential concentration risk.
Different prediction markets currently have different resolution mechanisms. If several major platforms eventually rely on the same oracle infrastructure, a failure affecting that infrastructure could potentially impact multiple venues at once.
The technology may be efficient, but greater efficiency can also create greater dependence on a common system.
The market that could expose the weaknesses
The real test for World will come when a market does not have a straightforward answer.
Take a contract asking whether a political leader will resign by a particular date.
At first glance, the question seems simple. But what happens if the official announces a future resignation? What if they are removed instead? What if they resign and later reverse the decision?
These situations show why the wording and resolution source matter so much.
The same problem applies to geopolitical events, political statements and other markets where the disagreement is not about the data itself but about what the data means.
Those are precisely the types of markets that have historically created the biggest controversies for prediction platforms.
World’s 150,000-market figure therefore needs to be examined more closely.
If most markets depend on clear, machine-readable information, automated settlement could be a strong fit.
If a large portion relies on subjective or contested definitions, the lack of a dispute mechanism could become a serious weakness.
Distribution could be World’s biggest advantage
The biggest threat to existing prediction markets may not be World’s technology. It may be distribution.
Kalshi and Polymarket have spent significant resources building their consumer brands through advertising, sports partnerships and corporate deals.
World began with access to a large existing wallet audience.
Users already using Phantom can access the product without going through the traditional process of opening a brokerage account, funding an exchange account or downloading another application.
That dramatically reduces friction.
But the same features that make World easy to access also create regulatory questions.
Skipping traditional onboarding can remove barriers for legitimate users, but it can also make compliance, identity verification and participant screening more complicated.
This creates an unusual competitive equation:
The architecture that makes World easier to use is also part of what makes its regulatory position harder to define.
What this means for traders
Anyone considering trading on World should pay close attention to the settlement rules rather than relying only on the market title.
Read the resolution source
The exact data source matters because there is no human review stage to reconcile an ambiguous market title with the underlying information.
If the market wording and data feed measure slightly different things, the specified resolution source is what ultimately matters.
Understand that settlement is final
There is no traditional challenge window.
Once the automated settlement process has run, participants cannot rely on an operator or dispute panel to reconsider the outcome.
Check postponed-event rules
Sports markets frequently involve postponed, abandoned or rescheduled events.
Before entering a position, traders should understand exactly what the contract says will happen in those situations.
Account for the lack of recourse
A non-custodial system can reduce traditional counterparty risk because the platform does not hold customer funds.
But there is another side to that equation: if something goes wrong outside the conditions covered by the code, there may be no centralized party responsible for correcting the problem.
What to watch next
Several developments will help determine whether World’s model works at scale.
The composition of its 150,000 markets will be particularly important. The more markets rely on objective, machine-readable data, the better suited they are to automated settlement.
The first major disputed settlement will be another critical test. The platform’s architecture has not yet been seriously tested by a large group of participants who believe an automated outcome is wrong.
Geographic restrictions are also worth watching, particularly because World has listed contracts involving sports, elections and monetary policy — areas already facing regulatory scrutiny.
Competitor adoption of automated settlement could also shape the industry. If platforms increasingly use automated oracles for straightforward price-based contracts while reserving human or governance-based resolution for subjective events, the market could eventually settle on a hybrid model.
Finally, trading volume matters more than the number of markets created. More than 150,000 markets demonstrates supply, but actual trading activity will show whether users are genuinely engaging with the platform.
A different model, not necessarily a better one
World’s approach does not automatically make it better or worse than Kalshi or Polymarket.
It simply moves the trade-off.
Kalshi offers human judgment, regulation and a formal path for complaints. Polymarket offers decentralized dispute resolution through an oracle mechanism. World prioritizes speed and automation by relying on data feeds and removing the dispute stage.
The central principle is straightforward:
The more discretion a system removes, the more correction it also gives up.
That may be an excellent trade for markets with objective outcomes.
For markets involving definitions, interpretation and unexpected real-world events, it could become a much bigger problem.
The success of World will ultimately depend on how well its automated resolution model matches the markets it chooses to list.
