On May 19, 2026, at a regular press conference in Beijing, Chinese Foreign Ministry spokesperson Guo Jia Kun confirmed that the leaders of China and the United States had engaged in "constructive communication" on artificial intelligence issues and agreed to initiate an intergovernmental dialogue mechanism, attempting to create a regulatory framework for this general technology amidst a long-term competition - however, specific timelines, topic scopes, and participating institutions have not been publicly announced. The dialogue's leaning towards security caution or industrial competition can only be speculated upon from fragmented information. On the same timeline, the London regulators reached down to the underlying financial infrastructure: the Bank of England, in conjunction with the UK's Financial Conduct Authority, proposed to extend the existing settlement system to operate "almost around the clock," explicitly framed by officials as paving the way for the tokenization of wholesale markets and RWA upgrades, yet details regarding the technical paths, implementation rhythms, and supporting regulatory texts are still undisclosed. Meanwhile, research institute SemiAnalysis released empirical data on AI outsourcing for nine real workflows, concluding that the "returns are both real and structural" after comparing token consumption with labor costs, providing capital markets and regulators with a new reference point regarding efficiency dividends - although these conclusions currently stem from a single source, and some samples remain to be verified. As AI compresses labor costs on an assembly line basis and nearly round-the-clock regulated settlement infrastructure nurtures a new generation of tokenized wholesale markets, the tension between China-US regulatory dialogues, UK infrastructure reconstruction, and corporate computational power ROI is elevating a core question: Will future technological dividends be locked within licensed and regulated systems, or will they continue to release spillover effects in the marginal areas?

China and the US at the Same Table Discussing AI: Regulatory Dialogue Officially Launched

Under the long-term tug-of-war of chip export controls and domestic computational power acceleration, the confrontation regarding artificial intelligence between China and the US has remained within the rhythm of "you impose restrictions, I fill in the gaps." On May 19, 2026, Chinese Foreign Ministry spokesperson Guo Jia Kun revealed that the two countries' leaders had engaged in constructive communication on AI issues, confirming their agreement to conduct intergovernmental dialogue on artificial intelligence, aiming to "jointly promote AI development and governance." This step shifts the competition from a war of words and unilateral lists to a trajectory with institutionalized temporal dimensions: the two sides are no longer just drawing red lines, but are reserving a formal table for potential future rule coordination. At the same time, Guo Jia Kun also clarified that there is currently no public timeline, topic scope, or participating institutions, indicating the regulatory dialogue has merely been lit up with the sign "coming soon," while the stage setup remains uncertain.

For cross-border AI companies, cloud service providers, and large model developers, this uncertainty is itself a clear signal: compliance boundaries may no longer be solely determined by unilateral export controls and local regulations but may be redrawn within the China-US dialogues. In the face of a fragmented global AI regulatory landscape, this intergovernmental dialogue may open a more predictable "compliance channel" for cross-border computational leasing, cross-border model deployment, etc., while possibly raising more coordinated and higher regulatory standards, requiring cross-border data and algorithm supply chains to simultaneously meet the "lowest common denominator" in both China and the US discourse. For other jurisdictions, the choice of the two artificial intelligence superpowers to sit down and discuss governance is itself a strong demonstration signal: who chooses to converge and who opts for differentiated regulation will directly impact their regulatory competitiveness in technology-intensive industries such as AI and tokenization. What truly determines the industry's direction will be whether this dialogue ultimately locks competition within rules or writes rules into competition.

AI Outsourcing ROI Empirical Tests: A New Reference Framework for Regulation and Capital

At the same time China and the US prepare to write AI into institutional dialogues, SemiAnalysis presented another set of "on-site data" capable of changing perceptions. This research institution, focusing on the semiconductor and AI industries, selected nine real workflows that included company research, financial report dissemination, meeting minutes mining, and financial data pulling, comparing the token consumption and labor costs of similar tasks, and concluded: in these scenarios, the returns on AI outsourcing are real and structured. In other words, in their samples, AI is not a one-time demonstration but can be decomposed into reusable and measurable production factors. However, this ROI calculation currently comes from a single source, and specific multiples and methodologies lack cross-validation, providing an enticing story to capital markets while also reminding regulators that it resembles an audit-basis sketch rather than an established fact.

Once this cost structure is laid out, the first mindset to be reshaped is that of roles in compliance, auditing, research, etc., which highly depend on professional labor: when internal calculations show significant cost reductions in AI outsourcing for company research or financial report interpretation, management will naturally inquire which processes can be outsourced to models without crossing regulatory red lines, and which steps must retain "human responsibility" to satisfy regulatory scrutiny. For auditing and compliance departments, high ROI expectations signal two new regulatory focal points: firstly, whether companies truthfully disclose the proportion of key analyses and reports generated by AI and their potential biases; secondly, whether using AI-generated research conclusions or financial interpretations exacerbates information asymmetries between insiders and external investors. As AI outsourcing costs are quantified as token bills, capital markets may be required to distinctly list "model dependency risks" and "algorithm outsourcing costs" in prospectuses and annual reports, while regulators will redefine what constitutes "reasonable AI use" and what constitutes systematic risks requiring additional disclosure and scrutiny by referring to empirical tests like those from SemiAnalysis.

Bank of England and FCA Extend Settlement Time to Facilitate Tokenized Assets

From a regulatory perspective, the "extension of settlement time" proposed by the Bank of England and FCA is not simply an operational duration extension, but an attempt to realign the central bank-backed settlement infrastructure with the capital markets and tokenized assets that are nearing round-the-clock operation. In official statements, this proposal is directly classified as "an important measure to promote the transformation of the UK's wholesale market towards tokenized finance," indicating that the regulators view it as a system-level foundation: one end connects with tokenized securities and RWA projects authorized by the FCA, while the other end maintains the central bank's usual risk boundaries. The current proposal's technical details, implementation timelines, and specific regulatory text have yet to be made public, leaving the market to anticipate future structures under the directional signal of "nearly around the clock": whether it will gradually extend the window or switch modes all at once becomes the first layer of uncertainty in the regulators' and institutions' games.

For the wholesale market, once the regulated settlement infrastructure approaches round-the-clock operation, the key constraints on tokenized assets, especially RWAs, will shift from "can it run on-chain" to "can it access the central bank's settlement pipeline." The Bank of England is responsible for infrastructure security, while the FCA delineates which tokenized projects qualify for access via its licensing system, creating a new regulatory boundary between 24-hour on-chain trading and traditional clearing systems: participants can obtain nearly uninterrupted funding and security settlements at the central bank settlement layer, while unauthorized crypto platforms remain firmly outside the system. For project parties, this means that tokenization is no longer a narrative of "skirting the old system," but rather aiming to strive for inclusion in the next round of infrastructure upgrades within regulatory-defined timelines and technical paths. Those who can first be included in this nearly round-the-clock settlement band will likely hold dual advantages in liquidity and compliance in the next stage of RWA competition.

Fight for Licenses and Channels: Who Can Use the UK's Tokenized Settlement Tracks

The first gate to connect with this new track still resides with the Bank of England and FCA. Traditionally, the core settlement systems regulated by the central bank and FCA serve only large banks, clearinghouses, and regulated market infrastructure, and now, under the “nearing around-the-clock” proposal, priority still goes to those already on the licensing list. The UK's tokenized finance strategy inherently tends to make regulated financial institutions responsible for issuance, custody, and settlement duties, rather than directly opening interfaces to unlicensed crypto platforms; coupled with the FCA's registration system and anti-money laundering requirements, unregistered platforms are restricted from providing certain tokenized products to UK investors, making it naturally challenging to be included in this central bank-level settlement track.

Once the near-round-the-clock regulated settlement infrastructure is implemented, the first to gain benefits will be the regulated participants in the wholesale market - banks, clearing institutions, licensed exchanges, and custodians - along with RWA projects and tokenized assets already falling under the purview of the UK regulatory framework. They can attempt on-chain bonds, tokenized government bonds, and other structures within the safety boundary of central bank accounts, while unregulated DeFi protocols and offshore unlicensed platforms remain excluded from the track, only able to seek space between "offshore" liquidity and limited internal access within the UK. This "official tokenized track" directly rewrites the global product map for RWA and on-chain bonds: projects seeking UK institutional funding must first design forms acceptable to the FCA's licensing system and central bank infrastructure, with compliance arbitrage shifting from "running on-chain" to focusing on "playing with licensing combinations and access qualifications." The true dividing line is no longer who dares to be more aggressive, but rather who can survive longer within the gaps defined by regulations.

A New Order of AI and Tokenization under Global Regulatory Competition

From the intergovernmental dialogue on artificial intelligence agreed upon by China and the US to the near-round-the-clock settlement proposal put forward by the Bank of England and FCA, superficially, these are two lines with different discursive systems, but fundamentally they point towards the same thing: redefining the feasible paths for technology and capital through regulatory frameworks. Currently, China and the US have only provided a framework commitment for "initiating dialogue"; specific timelines, topic scopes, and participating institutions have not been made public, meaning that any extrapolation regarding model liability, computational boundaries, or even cross-border data use must occur amid information gaps; on the UK side, both the central bank and FCA have merely released directional signals about expanding settlement infrastructure to serve the wholesale market's tokenization and RWA, with detailed technical plans and implementation timelines yet to be disclosed, leaving substantial policy design and technical development cycles before true implementation. Meanwhile, the optimistic productivity data provided by SemiAnalysis represents only one side of the market reference, and regulators have yet to officially acknowledge it, likely subject to reevaluation under a prudent regulatory perspective in the future. For crypto projects, this suggests that the single "regulatory safe haven" era is shrinking: products targeting UK institutional funding must be anchored against the FCA's licensing and central bank infrastructure access standards, reserving design space for compliance, data governance, and on-chain transparency constraints; for licensed institutions, when laying out cross-border AI and tokenization operations, they must simultaneously interface with fragmented AI governance rules and tokenized financial regulations in places like the UK, facing the costs of overlapping or even conflicting regulatory requirements; for end-users, compliance pathways will increasingly rely on three choices: "which jurisdiction, through which licensed institution, to access which types of tokenized assets." In this evolving regulatory map, true competition is no longer about who first embraces new technologies but rather about who can preemptively arrange compliance options across multiple jurisdictions within the incomplete regulatory fine points, maintaining the qualification to continue participating and benefiting under the dual orders of AI and tokenization.

Join our community to discuss and become stronger together!
Official Telegram community: https://t.me/aicoincn
AiCoin Chinese Twitter: https://x.com/AiCoinzh
OKX Benefits Group: https://aicoin.com/link/chat?cid=l61eM4owQ
Binance Benefits Group: https://aicoin.com/link/chat?cid=ynr7d1P6Z

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。