The World Cup provides a practical scene for watching how crypto wallets are changing. Questions that fans discuss every day, such as whether a team can qualify or how championship odds are moving, become priceable and tradable behavior on prediction markets such as Polymarket. When mainstream Web3 wallets integrate World Cup prediction activities, the result is more than a campaign around sports. Over a longer timeline, it becomes a lightweight entry point for users to begin interacting on-chain, in line with the discussion in World Cup carnival, prediction markets at the table: how Polymarket and others open a path toward mass adoption.
imToken’s World Cup-related experiment places AI Agents inside actual user scenarios. Its web-based agent and Discord agent are designed around concrete prediction needs, helping users complete betting transactions in a more natural flow. Users no longer have to perform every action inside a wallet app. They can participate through Discord or a web page, and then be guided by the agent back to on-chain interaction. This already resembles an early form of the Agentic Wallet: a future Web3 wallet may not be limited to a wallet app, but instead take the form of an AI-powered wallet presence that appears wherever users are acting.
From asset entry point to intent interpreter
For the past decade, the core questions addressed by wallets have been clear: where assets are stored, who controls the private key, and who signs the transaction. A user opens imToken mainly to check balances, transfer assets, interact on-chain and manage multi-chain holdings. In that phase, the wallet is best understood as an asset entry point and a signing entry point. The user knows what they want to do, and the wallet completes the final step.
The Agentic era changes that sequence. A user may not know at the start which button to press or which market to open. In a World Cup scenario, an ordinary user may not begin by saying they want to open Polymarket, find a particular market, judge the odds and execute a trade. They are more likely to ask what is interesting about tonight’s match, whether there is a market related to Portugal qualifying, whether the odds are already too low, or how to participate with a small amount. In the past, these questions were scattered across WeChat groups, social platforms and search engines. Each user had to assemble the information and then execute step by step. With an Agent involved, the interaction changes: the user expresses a broad intent, the Agent breaks it into a path, and the wallet turns that path into a series of on-chain actions.
This is not simply about adding a chat box to a wallet. The deeper shift is that the wallet moves from being a feature menu to being an intent interpreter. Older wallets ask users to decide whether they want to transfer, swap, stake or connect to a DApp. The next wallet interface can start from a natural-language goal. The World Cup is a useful entry point because it already has context, because users naturally want to express opinions, and because judgment is part of the experience. The Agent does not need to manage a complex asset portfolio from the start. It can first help a user find an interaction path inside one specific scenario, and then return final control to the wallet and the user.
Wallet entrances move beyond the app
The imToken web agent and Discord agent are clear examples of wallet capability moving into lighter entrances. They bring on-chain interaction into a campaign page, a World Cup context and a community conversation. Users do not necessarily need to open the app first, nor do they need to begin from a traditional DApp browser. They can discover the path through an Agent and then complete the on-chain portion when control returns to the wallet.
This means the boundary of the wallet is expanding outward. In the past, the entrance was easy to describe: open the app, enter the asset page, click a function and connect to a DApp. In the future, wallet entrances can be distributed across web pages, Discord, Telegram, AI chat boxes, event pages, developer tools and even lightweight wallet interfaces generated by users themselves. From this angle, the World Cup prediction campaign is not the core issue by itself. The more important point is that it lets the wallet stand in front of user intent in a more natural way.
If viewed only from within crypto, Agentic Wallet can be mistaken for a narrative label: AI helps users watch markets, find opportunities and execute trades, sounding like an extension of the last AI Agent wave. Mastercard’s launch of Agent Pay for Machines on June 10 placed the matter in a broader frame. Agentic interaction is no longer only a Web3 topic. It is also becoming part of the payment infrastructure conversation in traditional commerce.
Agent Pay and AI-dedicated cards offer a payment reference
Mastercard defines Agent Pay as a way for trusted AI Agents to participate in payments under user authorization. The design involves how an 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 completed with Agent assistance. These questions are highly similar to those faced by Web3 wallets. When AI only helps write copy, the cost of mistakes is usually manageable. Once AI participates in asset interaction, the issue changes: does it have permission, did it understand the user’s intent accurately, is the service it calls trustworthy, did the transaction exceed its boundary, and who bears responsibility if the result does not match the user’s expectation?
Mastercard’s answer is to redesign identity, tokens, authorization, risk control and dispute handling for trusted Agents inside the payment network. Domestic payment giants are also moving in the same direction. WeChat Pay is testing AI payment functions together with Tencent’s agent product WorkBuddy, including an AI-dedicated card inside WeChat Wallet. Based on disclosed information, the core of such products is not to let AI spend freely. Instead, they set boundaries for agent payments through stored-value limits, payment authorization scope and password verification. This follows the same logic as Mastercard Agent Pay: AI can participate in payment, but it must be identified, authorized, restricted and auditable.
The on-chain version of the same question
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. Because of this, Agentic Wallets cannot simply copy the traditional payment route. In traditional payments, users can rely on banks, card networks, merchants, dispute handling and risk-control systems. In the on-chain world, once a transaction is recorded, there is often no undo button.
The more efficient an AI Agent becomes, the more important the wallet becomes as the final safety boundary. A future wallet should not only let the Agent do things. It must ensure that the Agent only acts within the range allowed by the user. This is also why Web3 is a strong environment for discussing Agentic Wallets. Many discussions of AI Agents immediately drift toward automatic trading, automatic wealth management, automatic airdrop hunting and automatic arbitrage. Those directions are attractive, but for a wallet the hard problem is not automation. The hard problem is boundary awareness.
A wallet is not an ordinary application. If AI recommends the wrong song, writes the wrong paragraph or clicks the wrong page, the result is mainly a product-experience problem. In a wallet, if AI misunderstands an instruction, calls the wrong contract or grants an excessive approval, it can become a real asset risk. This connects with the question raised in Sign is more than signing: when an AI Agent signs for you, who still holds control?
The first-principles question for an Agentic Wallet is therefore not how many tasks AI can complete for users. It is how users know what the AI is doing. This is why imToken has repeatedly emphasized control over the past several years. From self-custodial wallets to multi-chain asset management, and then to AI co-creation and Agent exploration, the continuous thread is not simply that there are more features. The user must be able to understand, confirm and control their digital world.
In the Agentic era, that thread becomes more concrete. A wallet needs to show the user 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 tedious, but they form the basis for whether an Agentic Wallet can function in practice.
The power of an AI Agent comes precisely from turning complex processes into simple user experiences. A single user sentence can be broken into more than a dozen execution steps. That is good for usability, but it also creates a safety challenge. The longer the path and the more intermediate links it contains, the more the wallet needs to bring key checkpoints back in front of the user. A strong wallet interaction does not necessarily show more technical details. It translates complex transactions into language that the user can understand. The World Cup prediction activity is an event entry point, while imToken’s web agent and Discord agent are community entry points. Agentic Wallets are beginning from these small scenarios and moving into real use.