The World Cup offers a concrete setting for observing how Web3 wallets are changing. Whether a team advances, how championship odds move, and what fans discuss every day can all become priced and tradable actions inside prediction markets such as Polymarket. When mainstream Web3 wallets integrate World Cup prediction activities, the activity is not only about sports; over a longer timeline, it becomes a lightweight starting point for users to enter on-chain interaction.
World Cup predictions as a live test for wallet agents
In its World Cup-related exploration, imToken has started placing AI Agents inside practical usage scenarios. The agents it has introduced on the web and in Discord can help users complete betting transactions more naturally around specific prediction needs. Users are no longer limited to operating only inside a wallet app. They can participate in prediction markets through Discord, web pages and other contexts, while being guided back on-chain by the Agent in a smoother process.
This looks like an early version of the Agentic Wallet. The Web3 wallet of the future does not have to be confined to a single wallet application. It can exist as a more ubiquitous AI wallet form that appears wherever the user’s activity is already taking place. For the past decade, the central questions solved by wallets were relatively clear: where assets are stored, who controls the private key, and who signs the transaction.
For example, when users open imToken, they usually want to check balances, make transfers, interact with protocols, or manage multi-chain assets. At that stage, the wallet functions mainly as an asset entrance and a signing entrance. As long as the user knows what they intend to do, the wallet only needs to help them complete the final step.
The Agentic era changes that starting point. Users do not always begin with a clear button to click. In a World Cup context, an ordinary user may not think in terms of opening Polymarket, finding a specific market, evaluating the odds, and submitting a transaction. Their natural questions are more likely to be: what is interesting about tonight’s match; I am optimistic about Portugal advancing, are there related markets; are the odds already too low; if I only want to participate with a small amount, what is the process?
From feature menu to intent interpreter
In the past, answers to those questions were scattered across WeChat groups, social platforms and search engines. Each user had to assemble the information and then execute the process step by step. Once an Agent enters the flow, the interaction changes. The user only needs to express a broad intent; the Agent helps break down the path; the wallet then turns that path into a sequence of on-chain actions.
That is why the shift is not simply about adding a chat box to a wallet. The deeper change is that the wallet starts moving from a feature menu to an intent interpreter. Earlier wallets mainly waited for the user to decide whether to transfer, swap, stake, or connect to a DApp. A future wallet interaction can be more direct: the user describes in natural language what they want to complete.
The World Cup is a suitable entry point because it already has context, gives users something to express, and requires judgment. The Agent does not need to begin by managing complex asset portfolios, especially because complex asset management carries higher risk. It can first help users find an interaction path inside one concrete scenario, then return final control to the wallet and the user.
The web and Discord agents mentioned in imToken’s World Cup exploration are a typical example. They bring wallet capabilities into lighter entrances. A user does not have to open the app first, nor enter a traditional DApp browser first. Instead, within an activity page or a World Cup scenario, the user can find an interaction path through the Agent.
This means the boundary of the wallet is spreading outward. Previously, the wallet entrance was relatively clear: the user opened the app, entered the asset page, clicked a function, and connected to a DApp. In the future, wallet entrances can be distributed across more places, including web pages, Discord, Telegram, AI chat boxes, activity pages, developer tools, and even lightweight wallet interfaces generated by users themselves.
Payment networks provide another reference point
If the discussion stays only inside Crypto, Agentic Wallets can easily be treated as another narrative: AI helps users read markets, find opportunities, and make trades. Mastercard’s launch of Agent Pay for Machines on June 10 moved the discussion beyond Web3. Mastercard defines Agent Pay as enabling trusted AI Agents to participate in payments under user authorization, including how the Agent is identified, how it is authorized, how it is verified inside the payment network, and how merchants, issuers and users know that a transaction was assisted by an Agent.
This is highly similar to the problem faced by Web3 wallets. When AI only helps write copy, the cost of an error is usually controllable. Once AI starts participating in asset interactions, the questions change: does it have permission; does it understand the user’s intent accurately; is the service it calls trustworthy; does the transaction it initiates exceed the boundary; and if the result differs from the user’s expectation, who bears responsibility?
Mastercard’s answer is to redesign the identity, token, authorization, risk control and dispute-handling framework for trusted Agents inside the payment network. The signal matters because, while Agentic ideas inside Web3 still carry a technical imagination, a traditional financial giant designing payment infrastructure for Agentic Commerce places the topic into a more practical commercial context.
Domestic payment companies are also moving in the same direction. WeChat Pay is working with Tencent’s agent product WorkBuddy to test AI payment functions, including an AI-exclusive card in WeChat Wallet. According to the disclosed information, the core of this type of product is not allowing AI to spend freely. It is about setting boundaries for agent payments through recharge limits, payment authorization scopes and password verification confirmations.
This follows the same line as Mastercard Agent Pay: AI can participate in payments, but it must be identified, authorized, restricted and audited. Web3 wallets face the on-chain version of the same problem. The difference is that traditional payment systems emphasize networks, merchants, issuers and compliance responsibilities, while on-chain wallets emphasize private keys, signatures, authorizations, contract calls and user self-custody.
On-chain wallets need stronger boundaries
Because of that difference, Agentic Wallets cannot simply copy the path of traditional payments. In traditional payment systems, users can rely on banks, card networks, merchants, dispute resolution and risk control systems. In the on-chain world, once a transaction is recorded, there is often no undo button. The more efficiency AI Agents bring, the more important the wallet becomes as the final security boundary.
Future wallets therefore should not only make it possible for Agents to act. They must ensure that Agents act only within the scope allowed by the user. Many discussions about AI Agents quickly move toward extreme automation: AI automatically trades, manages wealth, finds airdrops or performs arbitrage. That direction is attractive, but for wallets the hardest issue is not automation. It is boundary awareness.
A wallet is not an ordinary application. In an ordinary app, if AI recommends the wrong song, writes the wrong sentence, or clicks the wrong page, the issue is mostly about user experience. In a wallet, if AI misunderstands an instruction, calls the wrong contract, or grants an overly broad authorization, the error can become a real asset risk.
The first-principles question for an Agentic Wallet is therefore not how many things AI can do for the user. It is how the user knows what the AI is doing. This is also why imToken has emphasized control over the past several years. From self-custody wallets to multi-chain asset management, and then to AI co-creation and Agent exploration, the continuous thread is not simply more functions. It is that users should still be able to understand, confirm and control their digital world.
In the Agentic era, that thread becomes more specific. A wallet needs to help users understand who the Agent is, which capabilities it can call, how long the authorization lasts, whether it can operate across DApps, when the user must confirm again, and whether the user can pause or revoke it with one action. These questions may sound procedural, but they form the basis for whether an Agentic Wallet can truly work.
The strength of an AI Agent comes from its ability to make complex processes simple. A single sentence from the user can be broken into more than a dozen execution steps. That is good for experience, but it also creates a security challenge. The longer the path and the more intermediate links it contains, the more the wallet needs to bring key decision points back in front of the user. World Cup prediction activities are an event entrance, while imToken’s web Agent and Discord Agent are community entrances. Together, they show how Agentic Wallets are already beginning from small, concrete scenarios.