On April 27, 2026, Meta's most important AI bet in Asia was abruptly halted. The Office of Foreign Investment Security Review working mechanism of China's National Development and Reform Commission determined that Meta's investment in the AI agency Manus posed national security risks, resulting in a ban that required the reversal of this acquisition, which had been completed in December 2025 for about $2.5 billion, and the restoration of Manus's assets in China to their original state. For Meta, this was not just a reversal of a transaction, but also a direct collision with a new wave of scrutiny regarding "key technology foreign mergers and acquisitions security review."
Within 24 hours, the flames of conflict spread from the national security review venue to the public square on the blockchain. On April 28, on the X platform, on-chain investigator ZachXBT published a post, naming the World (formerly WorldCoin) project's tokenomics design: a structure of low circulation and high valuation, which he analogized to the model from the SBF/FTX era, with more sensitive accusations pointed at its practice of "exchanging a small amount of WLD tokens for iris and other biometric data." Against the backdrop of the World project previously being scrutinized in multiple countries for privacy and data protection, this round of questioning rapidly escalated into a public debate about token models and data ethics—especially as he mentioned that verified WorldCoin accounts were being traded on the black market, implying hidden risks of account and data abuse.
On the same day, industry giants on the other end continued to advance their growth games at their own pace. On April 28, the Binance wallet announced the second phase of the Bitway (BTW) Booster activity rules: users must achieve 61 Alpha points to participate, and each participation will consume 5 Alpha points (according to a single source). Amidst the noise of public opinion questioning AI mergers and crypto projects, such a points threshold and incentive mechanism constituted the most mundane and overlooked aspect of the exchange ecosystem—it continually shapes user behavior and allocates traffic and attention.
In a brief two-day window, one side saw Chinese regulators halt the Meta–Manus acquisition in the name of national security, while the other side saw ZachXBT bring the WorldCoin model to the ethical judgment stage. Additionally, the Binance wallet continued to adjust the incentive parameters of the Booster activity. These three seemingly unrelated news threads intertwined: AI merger reviews, crypto token ethics, and the operational logic of the exchange ecosystem were forced into the same narrative coordinate system, unfolding a multi-threaded game about the boundaries of "technology, capital, and data power."
Meta's $2.5 Billion Acquisition Emergency Brake: The National Security Line Drawn to AI
For Meta, Manus was originally a piece of the puzzle leading to the "next-generation assistant"—this AI agency startup headquartered in Singapore, associated with China, was engaged in wrapping models into "agencies" and embedding them into specific business scenarios. In December 2025, Meta completed the acquisition of Manus for about $2.5 billion, seen as a step to intensify its commitment in the AI field through mergers and acquisitions: no longer just building models, but incorporating teams and technologies already operating in frontline scenarios.
The turning point occurred on April 27, 2026. The Office of Foreign Investment Security Review of China's National Development and Reform Commission issued a decision, indicating that Meta's investment in Manus posed national security risks. This prohibition order requested the reversal of this acquisition and the restoration of Manus’s assets in China. The official statement clearly indicated that this was not an ordinary business dispute, but the "veto key" pressed by the national security review mechanism regarding foreign mergers and acquisitions, also meaning that a transaction that had already completed settlement could still be "backtracked to zero" for national security reasons.
What followed was a complex "reverse engineering" process. Meta began preparing to withdraw this approximately $2.5 billion acquisition, and Manus’s equity and asset structure were forced to be dismantled in reverse. Early investors in Manus, including Tencent, Sequoia China, and ZhenFund, planned to cooperate with this withdrawal process, returning to the equity table; while earlier entrants like Benchmark were already receiving refund arrangements. Funds returned along the original path, and details about who exited at what price and under what structure were buried within contractual and confidentiality clauses, but the mere act of "money being returned" served as a public demonstration to the market.
The signal released by this case was not subtle: the national security line had already been clearly drawn over the AI industry. In recent years, countries have tightened scrutiny over foreign mergers in key technology industries, and this time, China invoked the foreign investment security review mechanism in the AI agency domain—which combines technology stacks and data resources—to impose a "ban" on a landmark cross-border technology acquisition. For AI companies like Manus that simultaneously connect Chinese and overseas capital, technology, and data, this means that any future cross-border mergers must pre-calculate the "discount rate" of national security reviews; for global tech giants like Meta, it serves as a reminder: AI is no longer just a story of technology and business; it has formally entered the narrative of national security.
From Manus to Silicon Valley: The Chill of Cross-Border AI Mergers
In the Meta–Manus case, "national security" was explicitly stamped on the veto of an AI merger for the first time. On April 27, 2026, China's National Development and Reform Commission's Office of Foreign Investment Security Review halted Meta's investment in Manus, requiring the reversal of this approximately $2.5 billion transaction completed in December 2025 and the restoration of Manus's assets in China to their original state. For the market, this was not just a decision but a high-held signpost: cross-border AI mergers are no longer just about antitrust or industrial policy issues; political and security thresholds have been raised to the center of the table.
The structure of this transaction was almost a textbook case of regulatory sensitivity. Manus belonged to the AI agency field, mastering AI-related technologies and data resources, located in Singapore, yet linked to China; Tencent, Sequoia China, ZhenFund, and other former Asian investors stood on one side, while Meta stood on the other, and regulatory agencies chose to draw an "unbreachable" line in between. Now, this line has materialized into a series of retreat arrangements: Meta was preparing to reverse the acquisition, Tencent, Sequoia China, and ZhenFund planned to cooperate in this withdrawal, and investors like Benchmark had already received refund payments in their withdrawal arrangements—this was not just asset recourse, but rather a forcibly halted global capital chorus.
The first to feel the chill may not necessarily be Meta, but those AI startups with ties to China or operating in China. In the past, they spoke of technology curves and business models in roadshows, but now had to insert a section called "national security acceptability" into investment memorandums. Seeking external financing or being acquired by international tech giants was previously a default growth path, now it's turned into a thin string filled with uncertainties:
● On one hand, even if headquartered in Singapore, the US, or other regions, as long as the equity structure includes Chinese investors or parts of R&D or data resources fall within China's jurisdiction, they could be regarded as "sensitive assets";
● On the other hand, even if the buyer is a multinational company like Meta, a transaction completed months ago could still be retroactively overturned, and the exit path is no longer a linear story but filled with retrenchments and reconstructions.
For investment institutions, this uncertainty will quickly reflect on terms and valuations: the discount for foreign merger exits has widened, regulatory risk clauses have been strengthened, and many projects originally aiming to "sell to global giants" have been forced to weigh between "becoming champions of the domestic market" and "accepting longer holding periods." This unpredictability is quietly changing the fundraising narrative for early AI entrepreneurs: after technology, market, and team, the fourth question has become—"Can your equity and data structure pass through the eye of multiple national security reviews?"
Meta is not an isolated case; it just collided with a global trend that has been slowly rising. In recent years, countries have broadly tightened scrutiny over foreign mergers in key technology sectors such as chips, cloud computing, and artificial intelligence. Safety reviews have shifted from a few cases to a normalized procedure. Meanwhile, Meta's model of acquiring AI technology and talent reserves on a global scale has been forced to confront systemic risk reassessment in this new environment:
● Every acquisition involving Chinese-related assets must incorporate "regulatory veto probabilities" and "capital losses when transactions are reversed" into their models;
● Tools originally designed to quickly acquire technology and teams, once vetoed, become risks of asset integration obstacles and prolonged returns.
For merger teams in Silicon Valley, the message conveyed by the Manus case is: a large new "gray area" has emerged on the global AI merger map. These areas are not forbidden zones but are indicative that any attempt to connect Chinese capital, data, or R&D resources must bear an additional national security price differential. What Meta needs to recalculate is not just the $2.5 billion withdrawal cost, but the risk weight of the entire strategy concerning Chinese-related assets and acquisitions—at what level of intersection does this transaction become unworthy of "betting" on a security review?
Returning to Manus, its core business is AI agency, companies of this kind depend on massive data to train models, mastering key resources such as algorithms, user behavior, and task execution links. When the National Development and Reform Commission clearly included such transactions under foreign investment security review and issued a prohibition decision, it effectively proclaimed: AI technology and data resources have already been viewed as security elements similar to traditional strategic industries in capital flows. Regulators are no longer merely concerned with the direction of capital; they are beginning to interrogate: What data is used for model training? To what extent can algorithms be utilized by military, intelligence, or critical infrastructures? Once the answers cannot be safely explained within established boundaries, "prohibition" becomes a natural conclusion under cautious logic.
Capital desires global movement while national security attempts to delineate boundaries; AI perfectly stands in the narrow gap of the two. The turning point of the Meta–Manus transaction has only made this tension more clearly exposed: the story of cross-border flows of technology, data, and talent is no longer solely about innovation and efficiency, but now also entails a layer of electrified regulatory barriers. Next, when the focus shifts from AI mergers to the crypto world, debates surrounding biometric data, on-chain assets, and token design will also be placed within the same framework of questions— in an era increasingly defining openness in terms of "safety," which capital and data are allowed to cross borders and which will be pressed to pause.
WorldCoin Under Fire Again: The Ethical Black Hole of Iris for Tokens
When the question shifted from cross-border merger approval halls to on-chain accounts and biometric databases, the protagonist in the story changed to another name—World (formerly WorldCoin), and its associate Sam Altman.
The basic narrative of this project is simple: with a vision of a "global identity network," it brings people to a terminal, confirms "you are a real and unique person" through iris scanning, and then issues a token called WLD. For many participants, the process is distilled into a slogan—"scan once with your eyes, receive a little coin." Regardless of how cutting-edge the technology is packaged, its essence remains to use token incentives to exchange for biometric data, but the subjects are no longer email addresses or phone numbers, but irises, which are difficult to change and tied for life.
On April 28, 2026, on-chain investigator ZachXBT posted a long thread on the X platform, shining the spotlight back on this project. The angle he chose was not the grand narrative of "AI + On-chain Identity," but the least dignified part: token economics and vulnerable populations.
In his description, the token design of WorldCoin is summarized as a "low circulation, high valuation" model—most tokens are locked in the hands of project parties and early stakeholders, with limited portions actually circulating on the market, yet the overall valuation is inflated highly. ZachXBT publicly compared this structure to the practices of SBF/FTX: using tight circulation and narrative to raise paper valuation, creating an illusion of something that "cannot go down" in the secondary market. For a project touting its mission as constructing identity infrastructure "for all of humanity," such a comparison is arguably a particularly harsh indictment.
More shockingly are his accusations regarding user acquisition methods. ZachXBT claims that WorldCoin distributes small amounts of WLD to participants in low-income countries in exchange for biometric data such as irises. In other words, among populations disadvantaged in both income and information, minimal on-chain incentives leverage the collection of extremely sensitive and irreversible data. Once this exchange is priced as "a few tokens," it is no longer merely a transaction, but rather a form of structural oppression: the project party takes away data assets that can be infinitely replicated, analyzed, and monetized, while those being scanned surrender nearly irretrievable bodily identifiers in an asymmetric information environment.
ZachXBT also pointed out in the same series of posts that verified WorldCoin accounts are being traded on the black market, meaning the identity system, originally designed for one person, one account, has been sold as a tradable asset package. This scenario is highly ironic: a network claiming to solve the "distinction between robots and humans" has its own accounts commodified in the dark. The moment accounts are transferred, whether the original holders are aware and how the iris data will be further utilized become unanswered questions.
These accusations quickly ignited public opinion also because they touched on vulnerabilities already exposed by WorldCoin—privacy and data protection. During its widespread promotion, the project faced regulatory scrutiny in multiple countries for collecting biometric data through iris scanning. On the surface, this concerns procedural issues—"whether data was collected in compliance, whether sufficient disclosure was provided;" at a deeper level, it reflects a systemic vigilance by regulators regarding "who is massively siphoning domestic citizens' biometric information, where this data will be stored, and for what models and business purposes it will be used."
When ZachXBT’s post went viral on April 28, 2026, the timeline happened to coincide with another event—just one day after the announcement of China's state security decision to halt Meta's acquisition of Manus. This chronological proximity amplified a common question: whether it’s AI mergers or crypto projects, when they touch upon the convergence of cross-border capital flows and sensitive data collection, how much "space to just get started" can still be relied upon?
For WorldCoin, the risks are no longer just about token price fluctuations or the enforcement actions of a single country, but a more challenging variable—market trust. In an identity system, if users start to suspect that their irises are being exchanged for a token that could be worth zero at any moment, rather than a seriously regarded long-term commitment; if regulators begin to view the project as a potential conduit for data breaches, rather than a pioneer of "inclusive technology," then the network’s fractures at the ethical and compliance levels will be difficult to bridge with PR rhetoric.
Rather than say WorldCoin has again come under attack, it is more accurate to say it faced an unavoidable question that would eventually be posed: in an increasingly secure-oriented world, pricing human biometric characteristics with tokens raises the question of whether it is technological progress or a beautifully packaged ethical black hole.
Exchanges Continue to Operate: Binance Wallet’s User Acquisition Game Goes On
On the same day World was pushed back into the limelight of public opinion, another timeline quietly unfolded. On April 28, 2026, the Binance wallet announced the second phase rule updates for the Bitway (BTW) Booster activity: the participation threshold was set at 61 Alpha points, and each participation would consume 5 Alpha points—this specific value and consumption setting came from a single source. Rather than view this as a simple activity update, it is more accurate to say it's the platform's periodic refueling of the "traffic machine" amid regulatory noise.
The Booster model essentially serves as a "task hall" within the exchange ecosystem. It doesn’t directly reward users in “recharges” or “trades,” but rather directs them towards specific chains or applications through intermediaries like points, whitelists, and interactive tasks: completing an interaction on a certain chain, executing an operation within a specific DApp, and then providing user feedback through task completion rates, raffles, and point accumulation. For the platform, this is a more refined operational system than simply issuing subsidies—it can guide traffic to projects it wishes to support while avoiding turning subsidies into one-time "freebie grabbing" opportunities.
Within such a system, an entry barrier like the 61 Alpha point requirement (from a single source) serves as the first filtering sieve. Only users who have accumulated sufficient interactions and task records in the previous phase qualify to enter the new round of Booster. The threshold keeps "passersby" at bay while keeping “those who have already invested time and effort” inside; and each participation requiring 5 Alpha points to be consumed (also from a single source) forms the second mechanism—compelling users to weigh between participation and retention, continually returning to the task system to "make up" their points.
This is a classic example of behavioral design: thresholds delineate an "inner circle" boundary, while consumption sustains user engagement. To avoid being ousted from the inner circle, users will continually complete new on-chain interactions, attempt more projects; project parties will gain traffic and initial user experiences through being included in the Booster task list. For exchanges, the wallet becomes the starting point for all these interactions—it is no longer just a tool for asset storage, but the primary interface for users with the entire ecosystem, and daily active users, retention, and task completion rates are constantly magnified in this points game.
Stepping further back, an ironic timing contrast emerges. On April 27, the Meta–Manus acquisition case was halted under national security review; on April 28, ZachXBT pushed the World project into the ethical and compliance spotlight; and on the same day, the Binance wallet continued to update the Booster activity announcement, guiding users to check-in on-chain, accumulate points, and consume points as usual. These three messages almost overlap on the timeline but belong to entirely different worlds: one side involves a stringent audit of capital, data, and control; the other side is a continuous effort to manage user attention and participation.
This contrast offers a specific cross-section for understanding the current dynamics between AI and the crypto industry. Regulatory authorities redraw boundaries with prohibition decisions and compliance reviews, while platform parties fortify their traffic pools through point thresholds and task incentives. The Meta–Manus acquisition has been compelled to withdraw, the World project has been nailed to the cross of ethical inquiries, and platforms like the Binance wallet maintain user habits and market momentum through activities such as the Booster. This isn't simply about "who is right and who is wrong," but a reality structure: regulation and the market are like two parallel tracks, advancing in different directions at the same time, while the update of the Bitway Booster serves as a clear footnote of the simultaneous operation of both tracks.
The Eve of AI and Crypto's Game: Parallel Red Lines and Gray Areas
Re-aligning the timeline, in December 2025, Meta acquired Manus for approximately $2.5 billion, attempting to bring in AI agency capabilities and data resources through a cross-border merger; less than a year later, on April 27, 2026, China’s National Development and Reform Commission's Office of Foreign Investment Security Review vetoed this investment, citing national security risks, mandating the transaction's reversal and restoration of Manus's assets in China, while Meta and former Asian investors such as Tencent, Sequoia China, and ZhenFund pivoted to withdrawal and refund arrangements. The next day, ZachXBT brought the World (formerly WorldCoin) project to the forefront on X, alleging it exchanged iris scans for WLD tokens, adopted a low circulation high valuation design, and likened it to the SBF/FTX model, while also indicating verified accounts were being traded on the black market; meanwhile, the World project had already faced privacy and data protection regulatory scrutiny in multiple countries. On the surface, one involves a halted foreign merger while the other pertains to a token model under fire, but the underlying direction is highly consistent: key technologies and data assets have been integrated into the lens of national security, biometric and user data have been elevated to global focus in privacy protection.
In this adjacent 48-hour span, cross-border capital, token design, and data processing were all placed under the magnifying glass. The Meta–Manus case falls under the backdrop of tightening scrutiny over foreign mergers in key technology industries, with artificial intelligence and other fields regarded as national security infrastructures; the World project, meanwhile, faced inquiry over its practice of distributing small amounts of WLD to low-income countries in exchange for biometrics like irises, raising questions of whether it crossed the line by exploiting information asymmetry and regulatory gaps. AI and crypto projects are often inherently cross-border—the sources of funding, equity structures, token distribution, and user demographics straddle multiple jurisdictions, compounded by algorithmic capabilities and sensitive data processing, easily entangled by intersecting national security scrutiny, financial regulation, and privacy protection rules; any tightening of a single line can pull the entire structure.
However, the market did not hit the pause button alongside the timeline. On April 28, 2026, the Binance wallet updated the rules for the Bitway (BTW) Booster activity as usual: establishing a participation threshold of 61 Alpha points and stipulating that participating once would consume 5 Alpha points (according to a single source). Amid the public discourse concerning the disassembly of Meta–Manus and the questioning of risks concerning the World project’s accounts and data abuse, such activities remind market participants—that while regulation is drawing red lines, platforms are still maintaining user growth and on-chain operations through incentive designs; the gray area has not vanished, only its cost and uncertainty have risen.
Within such a structure, project parties are no longer just "technology entrepreneurs,” but must simultaneously act as engineers, compliance officers, and public relations managers. AI companies like Manus, which possess AI technology and data resources, must foresee any potential transaction reversals prompted by security reviews when accepting foreign capital; projects like World, if they continue to rely on biometric data and complex token economics, must explain their data usage boundaries and economic incentive logic across multiple jurisdictions. It becomes increasingly challenging to maximize any one of technology iteration, compliance obligations, and public trust separately. Investors are similarly affected: post-2026, relying solely on valuation, liquidity, and narrative heat for decision-making is evidently insufficient; national security reviews, privacy regulations, and cross-border capital controls must become integral components of asset pricing.
No one can provide a definitive answer on what regulatory paths crossover projects between AI and crypto may take in the future. A more unified framework that packages algorithmic abilities, capital structures, and data governance into a single set of rules might emerge; it is also possible to see "sandbox" or industry self-regulatory approaches allowing certain high-risk trials to proceed within manageable boundaries; the more realistic possibility is a long-standing regulatory puzzle—where different nations draw their own lines between national security, financial order, and privacy protection, while project parties search for feasible space within the gaps. The Meta–Manus halting, the World project being under inquiry, and Bitway Booster continuing its operations are merely snapshots on the eve of the game; the real script has yet to be written. For those standing at this intersection, the more pressing question is not "Will regulation come?" but rather when both red lines and gray areas exist simultaneously, how to navigate the uncertain borders and determine what roles to play.
Join our community, discuss together, and become stronger!
Official Telegram community: https://t.me/aicoincn
AiCoin Chinese Twitter: https://x.com/AiCoinzh
OKX benefit group: https://aicoin.com/link/chat?cid=l61eM4owQ
Binance benefit 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。
