Bancor AI Liquidity Mining—A DeFi protocol absorbs deposits under the guise of AI market-making before suffering a governance attack and executing a rug pull

Who Gets Targeted

The victims are mostly DeFi novices lacking knowledge of crypto asset governance mechanisms, retail investors interested in artificial intelligence concepts, and liquidity providers attracted by high annualized yields. Their ages are concentrated between 20 and 40, mostly distributed in East and Southeast Asia where crypto communities are active. These investors are prone to believing propaganda that 'AI market-making algorithms guarantee steady profits without losses,' neither understanding the token holding distribution and contract permissions of governance tokens, nor chasing annualized yields far exceeding market levels with a speculative mindset, treating 'having passed a security audit' as a security guarantee. Typical losses include deposited USDT and ETH becoming unwithdrawable, token prices instantly dropping to zero, and 230,000 members losing their life savings when similar AI quantitative Ponzi schemes collapse (media estimates not independently verified). On-chain funds are extremely difficult to recover after being transferred via mixers and cross-chain bridges.

骗局怎么运作

• Attracting Users: Project teams heavily advertise on social media and crypto communities, claiming their DeFi protocol adopts AI market-making algorithms capable of achieving higher annualized yields than traditional liquidity mining, while promising that smart contracts have undergone security audits, attracting users to deposit crypto assets.
• Absorbing Funds: After being attracted by high yields, users deposit mainstream crypto assets such as USDT and ETH into designated liquidity pools of the protocol. The project team further expands the scale of deposit absorption through airdropped tokens and referral rewards, creating a short-term fund siphon effect.
• Executing the Attack: Once the liquidity pool reaches a certain scale, insiders or external attackers exploit vulnerabilities in the decentralization of governance token voting rights, initiating malicious governance proposals and quickly passing them to transfer pool assets to anonymous addresses, or directly using backdoor functions in smart contracts to withdraw funds.
• Project Collapse: After the attack occurs, the project team shuts down the official website, social media, and communities, token prices instantly drop to zero, users are unable to withdraw funds, and assets are completely swept away. Victims can only track fund destinations through on-chain records, but recovery remains difficult.
• Fund Transfer and Laundering: Attackers or project teams split pool assets through cross-chain bridges and mixers into multiple anonymous wallets, and then cash them out through over-the-counter channels on centralized exchanges, cutting off the tracking path of on-chain funds.
• Crisis Management and Secondary Harvesting: Procrastinating withdrawals and pacifying communities under the pretext of 'contract upgrades,' 'liquidity migration,' or 'system maintenance,' while simultaneously inducing users to migrate remaining assets to new pools or add margin, harvesting the final batch of liquidity before completely losing contact.

红旗信号（看到这些快跑）

• 🚩 Exaggerated AI Promises: The project team claims that its AI market-making algorithms can stably generate ultra-high returns, but has never disclosed algorithm code or backtesting data, with vague technical whitepapers lacking verifiable on-chain proof.
• 🚩 Lack of Audits: Smart contracts have not undergone independent audits by well-known security firms, or only display forged audit reports, preventing users from verifying code security.
• 🚩 Opaque Governance: Governance tokens are highly concentrated in the hands of a few addresses, with voting rights distributed extremely unevenly, leaving the community unable to effectively check malicious proposals.
• 🚩 Historical Issues: The background of the project team is ambiguous, core members are anonymous, or they are associated with past rug-pull projects, yet they deliberately evade this information during external publicity.

真实案例

• In July 2026, a DeFi liquidity pool lost $912,000 due to an oracle attack, and the related token Balance Coin plummeted by 99%. Attackers exploited price manipulation vulnerabilities to drain funds from the pool.
• In 2026, the AI quantitative cryptocurrency platform OneAgent was exposed as a hermit crab-style scam, with its operating team linked to the defunct Jubi exchange, and assets deposited by users were gradually transferred.
• In 2026, an AI quantitative Ponzi scheme named ANT.FUN collapsed, with its token ANB plummeting from $0.19 to $0.00000020, a drop of 99.99%, wiping out the principal of 230,000 members. Previously, its predecessor, the ANT community, had multiple fraud records. (Source: [https://www.corg.cc/article/19380.html](https://www.corg.cc/article/19380.html))
• In November 2021, the exclusive token 'Squid Game' (SQUID) on the gaming platform, capitalizing on the popularity of the hit drama Squid Game, staged a classic liquidity rug pull: the issuer suddenly withdrew funds and absconded after the coin price was hyped up to $2,861.8. Within 5 minutes, the price plummeted to $0.0007, a drop of over 99.99%, trapping about 40,000 token holders, and the issuer remains anonymous and out of contact to this day. (Source: [https://m.thepaper.cn/newsDetail_forward_15205825](https://m.thepaper.cn/newsDetail_forward_15205825))
• In October 2021, AnubisDAO, a DeFi project claiming to be a fork of OlympusDAO, raised approximately $60 million in ETH during a token sale, after which the liquidity pool funds were transferred to another address 20 hours after the sale began. Investors accused the project creators of running away with the funds, and the whereabouts of the involved funds remain unknown to this day. (Source: [https://decrypt.co/84924/anubisdao-investors-lose-60-million-in-alleged-rug-pull](https://decrypt.co/84924/anubisdao-investors-lose-60-million-in-alleged-rug-pull))

Official Stance

• Office of Investor Education and Advocacy of the U.S. Securities and Exchange Commission (SEC) (jointly with the North American Securities Administrators Association NASAA and the Financial Industry Regulatory Authority FINRA), January 25, 2024, published 'Artificial Intelligence (AI) and Investment Fraud: Investor Alert' (Source: [https://www.investor.gov/introduction-investing/general-resources/news-alerts/alerts-bulletins/investor-alerts/artificial-intelligence-fraud](https://www.investor.gov/introduction-investing/general-resources/news-alerts/alerts-bulletins/investor-alerts/artificial-intelligence-fraud))
• Chongqing Regulatory Bureau of the China Securities Regulatory Commission (CSRC), August 28, 2026, published 'Risk Warning on Guarding Against Illegal Financial Activities Engaged in under the Name of RWA' (Source: [https://www.csrc.gov.cn/chongqing/c106400/c7655023/content.shtml](https://www.csrc.gov.cn/chongqing/c106400/c7655023/content.shtml))

How to Protect Yourself

• ✅ Research Project Background: Before depositing funds, check the project team member information, historical project records, and community feedback to avoid providing liquidity to anonymous teams or platforms with suspicious backgrounds.
• ✅ Understand Protocol Mechanisms: Carefully read smart contracts and governance rules to confirm whether a single address controls the majority of voting rights or contains backdoor functions, and refuse to participate in protocols with highly concentrated governance.
• ✅ Use Security Tools: Check token holding distribution, fund flows, and contract change records through blockchain explorers and contract analysis tools to promptly detect abnormal transactions and risk signals.
• ✅ Diversify Investments: Do not invest all crypto assets into a single liquidity pool, control the scale of a single investment, and set reasonable stop-loss strategies to avoid total loss of principal.