On July 10, 2026, the GenLayer Foundation took the lead at the forefront, pulling together 27 crypto and Web3 institutions, including OKX, MetaMask, and Matter Labs, to announce a name that sounds more like judicial reform than a new protocol—Internet Court. Its official positioning is not to recreate a blockchain but to provide an “Internet Court”-style arbitration mechanism for trading disputes between AI Agents. The key lies in the list of participants: when leading trading platforms and wallet infrastructure players personally get involved, this design, which could originally be seen as a technological experiment, instantly gains the potential to evolve into industry-level “quasi-judicial” rules. David Riudor, co-founder and CEO of GenLayer, candidly stated their starting point: if Agents complete negotiations, contracts, and payments at machine speed, then the arbitration mechanism must also possess machine speed. Otherwise, any post-hoc arbitration merely serves to redo procedures for uncontrolled automation. This collides with the current reality—traditional judicial processes are measured in months, while dispute resolution paths for platform user agreements often rely on tickets and manual review, all of which struggle to accommodate machine-to-machine contract disputes completed in seconds or even milliseconds. The question of who will underwrite this risk is precisely what the Internet Court aims to intervene and rewrite concerning platform responsibilities.
Legal Absence in AI Agent Contracting
In today’s on-chain business, AI Agents are no longer just “recommendation algorithms” but actual parties to transactions: they can autonomously negotiate prices and terms with other Agents based on preset strategies, initiate contract calls on-chain, complete signing and payments, and directly form financial positions and risk exposures. What is entrusted to Agents is often automated financial and commercial tasks—from pricing, matching, to settlement—all completed between machines at second-level, or even millisecond-level speed, with humans only setting authorization limits and strategy parameters at the beginning, but struggling to keep track of each individual “contract signing.”
However, existing courts and arbitration mechanisms are designed for human-to-human contract disputes: filing, evidence presentation, hearings, and decisions take at least several months, and often several years, heavily relying on manual review and participation by the parties involved, making it even more difficult to understand the real-time technical details on-chain. Nonetheless, transactions between Agents are instantaneously completed across multinational networks and multiple chains, rendering traditional issues such as judicial jurisdiction, applicable law, and evidence preservation into post facto challenges in the face of millisecond transactions. For regulators, such machine-to-machine trading primarily “resides” in two sets of mechanisms: one is platform user agreements and service terms, which pre-assign responsibilities via standardized contracts; the other is the technical reality that on-chain contracts, once executed, are difficult to revoke, relying on prior logic to bind results rather than post-hoc adjudication and enforcement. Regulation and legislation are still in exploratory stages regarding how to define the legal identity and liability of AI Agents, without finding a middle ground between “code execution is irreversible” and “rights remedy must be sufficient,” meaning that when Agents autonomously sign high-risk trades, traditional stop-loss valves and complaint channels are practically absent.
Internet Court to Address AI Disputes
The answer provided by GenLayer is to “build another layer of court” beyond the on-chain and platform. In the proposal published on July 10, 2026, the Internet Court is explicitly defined as an arbitration platform for AI Agent trading disputes—it is not a formal court established by any nation, but rather a protocol promoted by an industry alliance: Agents agree to a set of commonly applicable dispute resolution rules before trading, and when issues arise, they do not seek regulatory authorities or offline arbitration venues but “submit” disputes to this protocol, which provides a judgment within agreed rules. David Riudor, co-founder and CEO of GenLayer, stated directly: Agents complete transactions at machine speed, and there must also be a decision-making mechanism at machine speed; the Internet Court’s job is to become an arbitration platform that these Agents jointly depend on when disputes arise in trading.
From the perspective of responsibility, the Internet Court attempts to occupy a space traditionally “neglected” by platform terms and judicial vacuums—it is neither the underlying chain itself, executing transactions mechanically as long as conditions are met, nor a single application layer platform, resolving disputes solely through user agreements and black-box risk management; rather, it is inserted between the chain, payment channels, and applications as a layer of dispute handling that can be invoked by all parties. To this end, the protocol plans to integrate the Coinbase x402 payment protocol, ERC-8004 Agent identity standards, Google A2A Agent interoperability protocols, MetaMask Smart Accounts Kit, ERC-7710 delegation mechanisms, and x402 Facilitator, utilizing unified payment, identity, and authorization standards to connect different AI Agent business systems to the same arbitration and execution paths. Currently, without any national regulatory bodies or courts officially participating in its governance or explicitly recognizing its judicial enforceability, it resembles a set of “industry courts” preset by 27 institutions (including OKX, MetaMask, Matter Labs, etc.) for themselves and future Agent users, first drawing a boundary of rules at the technical and protocol levels, then observing how real-world regulations and judicial systems respond to this newly built boundary of responsibility.
Support from Coinbase x402 and ERC-8004
What is externally referred to as the “Internet Court” is actually a carefully assembled combination of underlying components. The payment layer is handled by the Coinbase x402 payment protocol and x402 Facilitator, responsible for initiating, routing, and settling funds between Agents across different platforms and chains using a unified format; the identity layer is managed by the ERC-8004 Agent identity standard, fixing each participating Agent transaction under a recognizable, referable on-chain identity; the Google A2A Agent interoperability protocol manages the transmission of instructions and states among Agents across different business systems and technical stacks; the MetaMask Smart Accounts Kit provides a toolbox for account abstraction and permission control, along with the ERC-7710 delegation mechanism, formalizing “who has delegated what rights to which Agent” into the protocol. The official terminology is “planned integration,” and the complete online state and technical details have not yet been disclosed, but these names themselves outline a technological backbone from payment to identity, and then to authorization and interoperability.
On this backbone, Coinbase x402 and ERC-8004 are intentionally shaped as “anchor points” for liability tracing: a payment initiated by an Agent and completed via the x402 protocol is no longer merely a transfer between one abstract address and another, but can be tied to an Agent identity defined by ERC-8004, traced back through delegation relationships to the underlying principal and authorization scope. This design allows the Internet Court to directly reference standardized payment trajectories and identity markers when adjudicating disputes, rather than “guessing at suspects” in a sea of nameless addresses. Meanwhile, Smart Accounts and ERC-7710 are responsible for structuring the permission delegation process, so that in the future, whether the platform itself or third-party compliance services plug into KYC, fund flow analysis, or authorization audit tools at these identity and payment anchor points, there will be an opportunity to embed traditional compliance checks into the Agent economy without breaking existing protocols, thereby granting this “machine-speed” arbitration framework a technical foundation to extend into the real-world liability system.
Who Recognizes Private Domain Arbitration After OKX and MetaMask Join?
When OKX, MetaMask, and Matter Labs appear on the list of 27 participants, the originally abstract “Internet Court” instantly gains tangible mechanisms. OKX controls vast accounts and capital inflow and outflow, MetaMask manages wallets directly aimed at end-users along with the upcoming Smart Accounts Kit integration, and Matter Labs participates in the design of the protocol and underlying standards. If these roles truly connect to the Internet Court within their respective products, then judgment will no longer be just a signature result on-chain, but could potentially be “automatically executed” on account permissions and transaction paths. From the perspective of industry dynamics, whoever first writes this mechanism into user agreements, risk control rules, and Agent business logic will possess factual discretion over AI Agent disputes.
For ordinary users, once a wallet or trading platform technically integrates the Internet Court, the direction of certain disputes may shift from “first find customer service, then go to court” to a procedure of “first having the protocol adjudicate, then having the platform execute in the background.” For example, the wallet side may adjust an Agent's authorization whitelist based on the judgment, block it from continuing to initiate payment instructions, or even change the subsequent fund flow according to the judgment result; on the trading platform side, it may output protocol integration during account freezing, withdrawal path choices, and on-chain delegation cancellations, allowing the “ruling” to be realized in the form of risk control rules. But an ensuing issue arises: this type of “private domain arbitration,” initiated by an industry alliance and devoid of any participation from national judicial authorities, is ultimately what in the eyes of regulators and courts? In the absence of any regulatory or judicial entity publicly acknowledging its judicial effectiveness, it may only be seen as a convention-based dispute resolution mechanism between platforms and users, and should a conflict arise with consumer protection clauses or mandatory laws (such as non-waivable litigation rights, or mandatory local laws), who will ultimately decide will determine whether the Internet Court is perceived as an internal compliance tool or a high-risk experiment challenging existing judicial orders.
A Few Industry Red Lines That May Be Redrawn
If agreements like the Internet Court are widely integrated by mainstream platforms and wallets, the “lines of platform responsibility” that were once rigidly defined in various “User Agreements” and “Smart Contract Disclaimers” might be moved to the protocol layer: disputes will no longer be “self-justified” by a single platform but will first be submitted to a cross-platform arbitration process, where Agents adhering to the ERC-8004 identity standard, ERC-7710 delegation mechanism, and Coinbase x402 payment protocol are reviewed under a shared set of rules. For platforms, this signifies a transition from “acting as the judge” to needing to explain why they opted for this “Internet Court” and the responsibility in cases of erroneous executions by Agents, contract calling anomalies, and where fault lies—whether with the platform, the Agent's developers, or any link in the user authorization chain. User complaint pathways will also be rewritten: shifting from previously filing support tickets or sending lawyer letters to a process where users first face a set of “machine-speed” adjudications written into the agreement and executed on-chain, and whether such results can still be overturned by offline courts will become a question that every contract deployer and every platform must answer in their risk control clauses.
From a regulatory and compliance perspective, the potential benefits of such agreements are also clear: unified Agent identity standards (like ERC-8004), delegation and authorization mechanisms (like ERC-7710), and combinations of payment channels make the commercial behaviors and dispute records between AI Agents inherently possess a traceable account system and data structure, facilitating post-hoc audits, on-chain evidence collection, and cross-platform collaborative investigations of suspicious fund flows. For regulators who have already prioritized AI, on-chain data, and cross-border capital flows, this “protocol-level arbitration” is easily integrated into future regulatory discussions as a sample. However, all these “benefits” currently remain at the level of technical and institutional design: the Internet Court has not yet disclosed a governance model, appeals mechanism, or selection of applicable laws, nor has any national regulatory authority or court publicly recognized its adjudicative force, making it seem more like an internally constructed dispute resolution infrastructure by industry alliances, while whether it can transition from “soft-standard arbitration” to being partially absorbed by judicial and regulatory systems will depend on subsequently disclosed rules, real dispute case implementations, and whether regulators across nations choose to acknowledge this new red line or actively clarify the boundaries.
Join our community to discuss and become stronger together!
AiCoin exclusive Hyperliquid benefits: https://app.hyperliquid.xyz/join/AICOIN88
AiCoin exclusive Aster benefits: https://www.asterdex.com/zh-CN/referral/9C50e2
On-chain Telegram community: https://t.me/AiCoinWhaleData
On-chain community: https://www.aicoin.com/link/chat?cid=N6OVMor5g
AiCoin on-chain Twitter: https://x.com/aicoinwhaledata
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
