On March 25, 2026, Beijing time, a key turning point regarding the controversy surrounding the Backpack airdrop witch judgments emerged. Previously, the project implemented extremely strict anti-witch rules during a large-scale token distribution, leading to the concentrated liquidation of numerous addresses marked as "studios" or "multiple accounts," causing collective dissatisfaction among avatar users, heavy players, and some studios. As public opinion continued to accumulate pressure within the community, the Backpack team announced that they would reevaluate complex cases and officially open an appeals channel while introducing a new rules framework known as the "Rule No. 3" — allowing users with no more than three accounts on the same device the opportunity to receive at least 50% of their tokens back, with some marginal cases even rumored to receive up to 100% back after review (according to a single Chinese source). This series of actions not only alleviated emotions but also refocused the contradictions on an old issue: in the Web3 world, which heavily relies on anti-witch risk control but lacks a mature identity system, how should project teams redraw the line between "prevention of fake accounts" and "protection of user rights"?

From Strict Crackdown to Re-evaluation: Backpack's Change of Attitude

The starting point of this wave at Backpack was a "high-pressure" witch hunt. To prevent airdrops from being exploited by studios and script players, the project adopted centralized and strong rules for judgment: dimensions such as multiple accounts on the same device, similar on-chain behavior, and address aggregation relationships were packed into a black-box algorithm, resulting in avatar users, heavy participants, and some small studios being swept together, directly losing all their token shares. For many retail investors who were mistakenly impacted, what they saw was not "anti-witch justice," but a "one-sided judgment lacking an appeal channel."

As the backlash grew louder, Backpack founder Armani Ferrante publicly stated, "Our team is reassessing more complex cases," which became the watershed moment for the project's change in stance. From "we have completed the judgment" to "complex cases need to be reassessed" indicated that the team recognized that the previous risk control rules had gray areas on the boundaries and acknowledged that technical screening could not resolve all disputes at once. The subsequent plan rolled out in three steps: first, to re-evaluate cases marked as complex or marginal; second, to formally open an appeals channel, allowing users to submit supplementary materials and usage scenario descriptions; and third, represented by "Rule No. 3," to publicly disclose certain black-box rules in simplified terms, providing an interactive reference point for the outside world.

This "about-face" is inevitably interpreted as public relations damage control: opening up an appeals process under public opinion pressure seems like a standard crisis management handbook. However, viewed from a more long-term governance perspective, it also exposed the project's initial underestimation of the "error correction mechanism" in anti-witch decisions — once penalties are too rigid without public standards and review pathways, technical risk control can quickly evolve into a crisis of governance legitimacy. This shift from strict account bans to revisiting appeals not only addresses the current reputation risk but also forces Backpack to incorporate "rule transparency and process appeal" into its long-term governance framework, rather than just a one-off technical parameter adjustment.

One Device, Three Accounts: The Boundaries and Gray Areas of Rule No. 3

In the re-evaluation and appeals mechanism, the most discussed topic is the so-called "Rule No. 3": according to several Chinese media outlets (such as Liudong and Xingqiu Daily), Backpack has set a key threshold — users with no more than three accounts on the same device have the opportunity to receive at least 50% of their tokens back. The superficial meaning of this clause is very intuitive: the project team attempts to draw a line for an "acceptable multi-account limit" in terms of device fingerprint dimension, distinguishing between obvious batch account abuse and potentially reasonable multi-account usage. However, upon implementation, the boundaries immediately become blurred: is it based on hardware ID, system fingerprint, browser fingerprint, or login behavior? Different fingerprint aggregation methods can lead to vastly different determinations of "the same device."

Regarding the refund ratio, brief information points to a hierarchical structure: the baseline corresponding to Rule No. 3 is "≥50% refund," and according to a single source from Xingqiu Daily, for some marginal cases, there exists the possibility of up to a 100% refund after review. This indicates that outside of the technical rules, there is an invisible track of "human discretion + individual case evaluation": individuals with complete narratives and evidence that can convince the review team that they are "not malicious witches" may receive an additional "full exemption" beyond the rules. However, the claim of a 100% refund currently comes from a single Chinese source, and a full disclosure of complete wording has not yet been seen from Backpack's official channels, leaving significant uncertainty regarding the reliability and applicability of the information, which requires investors and appeal users to remain cautious.

Rule No. 3 appears to be a relatively lenient threshold, but in real usage scenarios, it immediately exposes the dual space of misfires and exploitation. It is very common for mobile users to switch between "main accounts + small accounts + family and friends' accounts" on the same phone; in family settings, multiple family members sharing a tablet or computer to log into different wallets is also entirely reasonable; in internet cafes and shared office spaces, the cross-sectional IP and hardware environments are more complex. For these people, being identified as "multiple accounts on the same device" does not indicate that they are organizing a studio to abuse airdrops. On the other hand, professional studios can fully utilize device rotation, virtual machine isolation, and fingerprint disguise to split multiple accounts within several "three-account thresholds," ultimately forming an operational guideline of "as long as it's within three accounts, it's safe to abuse."

More troublesome is that the detailed regulations for Rule No. 3 have not been fully disclosed: for example, how to handle the conflict between historical usage data and single-claim behavior, how to view the differences between short-term logins and long-term bindings, and whether there will be further cross-screening in conjunction with on-chain trading patterns. These issues relate to compliance and communication risks: in the absence of a complete official statement, the community often has to self-interpret based on scattered information, easily forming a psychological gap where "the project speaks one way but acts another." If a large number of cases arise with statements like "I clearly only have three accounts but only got back X%," a new cycle of distrust will be triggered.

The Age-Old Dilemma of Witch Judgments: The Paradox of Technical Fraud Prevention and Misfires

The predicament Backpack currently faces is not an isolated case; it is a "witch paradox" that almost all Web3 projects encounter during airdrops and token distributions. Typical witch attack methods include: batch generating wallet addresses, utilizing scripts to complete tasks, centrally controlling private keys for unified claims, and then completing fund aggregation through several on-chain transfers. The project's risk control engine builds a multi-dimensional "suspicion scoring system" around device fingerprints, IP addresses, wallet aggregation relationships, interaction time distributions, task completion trajectories, etc., to extract clusters that exceed conventional usage patterns for centralized processing.

Regarding the technical solutions, common practices can be roughly classified into a few categories: first, device fingerprints and network environment, identifying potential studio clusters through browser fingerprints, hardware information, and IP ranges; the advantage is simple deployment and low cost, but the risk is heavy reliance on centralized service providers, which is inherently unfriendly to scenarios like internet cafes and family shared networks; second, multi-wallet aggregation analysis, judging the likelihood of "multiple accounts under the same master" through on-chain transfer paths, repeated interaction contracts, and addresses of funds' final aggregation; the advantage is transparency and auditability, but scripts have enough room to "wash behavior" through intermediary addresses and extended time; third, on-chain behavior profiling, concatenating interaction frequency, gas usage patterns, task completion order, etc., into behavior vectors, training models to differentiate "human players" from "script spammers"; the advantage is high granularity, but the downside is that features are easy to imitate and hedge, making it challenging to quantify the misfire rate in advance.

The issue is that, regardless of how the tech stack upgrades, it is difficult for projects to accurately distinguish between "full-time gold farmers" and "organized studio spammers" solely based on on-chain data and device information. A heavy participant might frequently interact across multiple networks, use automated tools for assistance, and manage several addresses simultaneously; from behavioral data, they are essentially no different from a small studio. Conversely, some resource-scarce small studios can fully masquerade as "ordinary players" by outsourcing devices, purchasing verified accounts, and faking usage trajectories. This creates a structural dilemma: the more stringent the technical risk control, the more challenging the balance becomes in weighing "presumption of guilt" against "misfires on individuals."

In this context, decentralized identity (DID) and on-chain reputation systems are seen as potential "solution clues" in future distribution mechanisms. Theoretically, by accumulating on-chain reputation, task completion history, and community participation over time, users can build a "transferable credit profile," gaining higher weight in new project airdrops rather than just relying on one-time address behavior. However, this path is also accompanied by clear ethical and privacy boundary issues: who will maintain this reputation data? Will it effectively form a "de facto on-chain credit institution"? Are users forced to make asymmetrical concessions between privacy and benefits? The recent controversy at Backpack has merely exposed these issues that should have been discussed in advance through the lens of "one-size-fits-all misfires" and "retrospective appeals."

The Anger of the Chinese Community and the Echo of Cross-Cultural Operations

Among all community reactions, emotions in the Chinese community have been particularly intense. After the initial version of the witch judgment was announced, many users who considered themselves "compliant participants" found themselves placed on the witch list, with the claim interface directly reset to zero or marked as non-compliant. On social media platforms, phrases like "misfired retail investors," "avatar users wiped out," "ordinary players are less safe than scripts," rapidly spread, while screenshots, appeal experiences, and comparison cases circulated like wildfire in groups and forums. Some users began to compile spontaneous appeal templates, while others directly questioned the project’s disregard for the usage habits of East Asian users in rule formulation, such as shared family devices and internet cafe environments.

In this process, the signals quickly released by the project such as "reassessing complex cases" and "opening an appeals channel" can hardly be said to be unrelated to the public opinion pressure from the Chinese community. While there is currently a lack of direct evidence proving that "the Chinese community drove this shift," the chronological alignment of intense concentrated feedback, dense media reporting, and project statements at least demonstrates one reality: for projects targeting a global user base, the Chinese market is no longer a marginal noise that can be overlooked, but an important force capable of influencing governance decision rhythms. (This correlation itself still requires verification and observation, rather than a definitive conclusion.)

On a deeper level, the systematic differences in information acquisition, modes of expression, and channels for safeguarding rights between the Chinese and English communities present further challenges. English community users are more accustomed to expressing dissatisfaction through GitHub issues, lengthy forum posts, and AMA questions to "institutionalize" complaints, while the Chinese community tends to rapidly amplify emotions through social groups and social media, employing screenshots for comparisons and public pressure. When official announcements primarily in English are released, Chinese users often rely on second-hand translations and media interpretations, leading to discrepancies in key wording (such as “may,” “could,” “under review”), further exacerbating feelings of "information asymmetry" and deprivation. Regardless of the ultimate outcome, Backpack's response at this time offers a textbook example for other globalized projects: when formulating anti-witch and risk control communication strategies, it is essential to consider cross-cultural contexts, information delays, and rights protection habits; otherwise, any technical rule may be "emotionally reinterpreted" during transmission.

Is the Appeals Channel a Band-Aid or a New Battlefield?

Adding an "appeals channel" as a remedy after technological risk control appears to be a quick fix, but the underlying game is far more complex than it seems. For users, the most critical issues are concentrated in three areas: who judges, how to provide evidence, and whether there is a space for secondary review and independent arbitration. If the final judgment remains entirely in the hands of the original review team and lacks publicly available assessment standards, then appeals merely add another layer of "explaining the situation to the same judge" to the original judgment, hardly alleviating distrust; if appeals require a substantial amount of KYC, device evidence, and usage scenario descriptions, many users may choose to abandon the process due to concerns about privacy and time costs. Secondary review and independent arbitration mechanisms directly relate to whether users believe that "individual grievances" can be corrected rather than dismissed as statistical errors.

At present, information regarding appeal pass rates and specific execution timelines is absent from public channels. The project has not disclosed any data on "what percentage of appeals may result in a reversal" or "how refunds will be executed in batches." In this state of opacity, users' evaluation of the credibility of refund commitments relies more on scattered success cases and media reports than on verifiable overall rules. Without a statistical perspective, market emotions can easily be influenced by extreme examples: isolated high-value refunds being amplified and shared may lead more people to feel "it might be worth taking a gamble on an appeal"; conversely, if a large number of unsuccessful appeal cases begin to accumulate, a new round of denunciation claiming "the appeals are merely a formality" will swiftly emerge.

Particularly, the information that "marginal cases may achieve full refunds after review" (from a single source from Xingqiu Daily) has a magnifying effect on expectations. For some users who have already been confirmed as witches, this undoubtedly opens a door to a "from 0 to 100" imaginary space — even if the probability is unclear, it still warrants the investment of effort to take a chance. This "betting on appeals" mentality elevates the appeals channel from a corrective mechanism to a new battlefield: some individuals package their usage stories and blur studio boundaries to solicit lenient judgments, while others mobilize through group pressure and emotional rallying to increase their "visibility" in the queue. From the project's perspective, if there is a lack of clear, consistent standards, individual refunds can become a new source of inequality: those who can tell better stories or mobilize public opinion are more likely to gain "excessive correction" in the marginal areas.

In the long term, this mechanism will directly influence subsequent token unlock emotions and project reputation curves. On one hand, opening appeals and partial refunds may mitigate angry emotions in the short term, creating a relatively less adversarial public opinion environment for future unlocks; on the other hand, if the appeals mechanism is widely regarded as "random outcomes" and "emotionally driven," then each new distribution or unlock will be reactivated by historical shadows, forming reflexive accusations of "you also said last time that there would be fair handling." What Backpack is currently doing is patching up after cracks have appeared in their reputation defense line, and whether this patching serves to "stop the bleeding" or "open new wounds" ultimately depends on their ability to provide a publicly understandable and acceptable procedural justice framework beyond individual warmth.

The Next Round of Anti-Witch Efforts: From Rules to Consensus

In reviewing the entire event, Backpack has walked a typical crisis path: it began with a strong anti-witch crackdown centered on technical risk control, trying to achieve a "cleaner" distribution environment through one-off cleansings; subsequently, after community backlash, particularly from the Chinese community, it shifted towards reassessing complex cases, opening appeals channels, and partially disclosing rules through avenues like Rule No. 3. The endpoint of this path has yet to arrive, but the phased results are already evident: the project has loosened its leverage on the "anti-witch" front, while the voices for "user rights" from the community have increasingly been included in decision-making, leading both sides to reluctantly push open a middle ground between technical algorithms and emotional appeals in the tussle of specific cases.

However, it is unrealistic to expect that one-time rule fixes can resolve witch controversies once and for all. Anti-witch efforts are inherently a dynamic game: once rules are announced, studios will optimize strategies around them, new evasion methods will force projects to raise thresholds, and misfires and distrust will repeatedly surface in this arms race. In this cycle, what is truly scarce is not more complex algorithms, but rather sustained transparency and data disclosure: including overall error rate estimates for witch determinations, numbers and proportion ranges of successful appeals, and anonymized descriptions of typical marginal cases. Only when the community can understand "what has overall happened" under relatively sufficient data can individual dissatisfaction cases avoid evolving into total denial each time.

Before a more mature decentralized identity and reputation system appears, what projects need to do is design a "appealable, correctable" distribution system under existing technological conditions: publicly disclose the approximate range of key indicators and judgment logic in advance to reserve institutionalized channels for correcting misfires, rather than constructing appeal entrances in a haphazard manner afterward; in the marginal areas where human discretion is needed, establish processes that minimize subjective judgments and reduce structural biases such as "the louder voice takes advantage." For users, it is equally necessary to remain calm when participating in airdrops and appeals: be wary of the information amplification effect from single sources, recognize the uncertainties behind expressions like "up to 100% refund," and rationally assess their compliance risks and time costs in the absence of pass-through rate data and execution timelines, rather than treating appeals as a gamble focusing solely on returns without considering costs.

The next round of anti-witch efforts is no longer just a cat-and-mouse game between project teams and studios, but how the entire industry can find a new balance between technical rules and community consensus. Backpack's recent adjustments may be an imperfect attempt, but it at least clarifies an essential fact: in the governance world of Web3, any impulse to resolve issues through black-box risk control ultimately must return to the public dialogue of "can it be explained, can it be corrected, and can it be understood."

Join our community to discuss and grow 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。