Community Update: Sybil Detection and Decision on $OM Allocation

As shared in our last community update, we detected multiple fraudulent accounts in our recent sybil investigations that were attempting to unfairly gain $OM. These accounts are not genuine members of our community, nor do they have an authentic and honest interest in the durability, sustainability and long term vitality of our ecosystem.

Today, we are sharing a further breakdown of the methodology our investigation employed. Wherein, we answer additional questions received since our community update. We are also providing a further update on the upcoming GenDrop.

The Methodology

To ensure fair $OM distribution and prevent sybil activity, multiple detection approaches were employed ahead of the upcoming GenDrop.

Our goal has always been to reward and protect community members who contribute to the long-term growth of our ecosystem, while preventing bad actors — namely bots and sybils — from exploiting it for profit, particularly to the detriment of genuine others.

The methodology involved engaging independent third-party firms, community reporting, and internal analysis. This included;

• Independent firms: External experts analyzed archival testnet node data implementing clustering algorithms, interaction graphs, and machine learning models amongst other methodologies. Wallets with 20 or more linked addresses were flagged, along with those exhibiting circular funding patterns, high-frequency bot activity, and other patterns.
• Community reporting: Informants provided additional insights, identifying patterns such as identical wallet creation dates, systematic GenDrop farming, and sequential wallet transactions.

For example: If Alice had 20 or more wallet addresses, and all were funded from a single wallet address onchain with the same pattern, then Alice and all her subsequent wallets were flagged as sybils.

• Internal analysis: Our team examined network-wide trends, linking wallets through repeated transactions and bridging activities. These findings were integrated into a composite scoring model to evaluate individual wallets.
• Validation process: Wallets that triggered more than one signal across different methods were flagged for review. Very strong indicators led to classification as a sybil, while those wallets that showed weaker indicators, underwent further scrutiny to minimize false positives.

The Findings

As a result of the methodology, the below was ascertained;

• 123,195 out of 217,351 wallets have been removed from eligibility. Please note, more accounts could have been classified as ineligible, however they were not.
• By failing to bridge and link their mainnet wallet addresses, currently 80,000 wallets remain unallocated.
• Reports with verifiable methodology were considered, and their results double checked. As a reward for uncovering 1,066,480 $OM linked to bot activity, 29 members of the community have been allocated 106,648 $OM for providing high-quality and valid sybil reports.

As an example, if a person submitted 100 wallets for review, these were first checked against a list of known sybil wallets. If 60 were already on this list, their reporting would not be rewarded, however the newly identified 40 would.

The reporter would then receive 10% of the total OM allocation for the sybil wallets they identified. Meaning, if each sybil wallet was worth 1 OM, the informant would receive 10% of the 40 correctly identified OM as their reward, so 4 OM.

This system ensured that informants were rewarded for uncovering new sybil wallets, not for reporting already-detected ones.

• Unfortunately, the self-reporting initiative was misused, with multiple reports falsely and maliciously targeting real users. We believe this was most likely done to reduce the total GenDrop allocation. To ensure that legitimate participants receive their GenDrop allocation, we have invalidated those self-reports where we could not reliably verify their accuracy.
• As a result, and in accordance with the proposal, 26,942,144 OM will be set aside for burning.

Next Steps

Given the magnitude of sybil activity detected and the value of the $OM, we believe it is in the best interest of the ecosystem for the community to decide the best course of action.

Next, we will share a governance proposal, whereby the community can vote on the future of these $OM. It is our hope that as many members of the community as possible partake.

Regarding sybil investigations, we will continue monitoring the ecosystem using external and internal measures. Any additional wallets identified as a sybil, will forfeit their allocations.

As shared previously, our investigations have been conducted to ensure those who’ve contributed to MANTRA’s growth to date and supported our work in bringing the world’s financial ecosystem onchain — are adequately and honestly rewarded.

We would like to thank all who have assisted to date.

Let’s keep building.

MANTRA.

About MANTRA

MANTRA is a purpose-built Layer 1 blockchain for real-world assets, capable of adherence to real-world regulatory requirements. As a permissionless chain, MANTRA Chain empowers developers and institutions to seamlessly participate in the evolving RWA tokenization space by offering advanced technology modules, compliance mechanisms, and cross-chain interoperability.

To request more information about MANTRA, or connect with a member of our team, you can do so here. To be the first to read about MANTRA, our product news, community updates and more, connect with us on Telegram, X, LinkedIn and Discord. We also recommend you visit our announcements page regularly.