Compound

Value

Flash

Tornado Cash

Balancer

<ul>
 <li><strong>An attacker drained $20 million from the BonkDAO treasury through a valid governance vote.</strong></li>
 <li><strong>The takeover required buying roughly 1% of the BONK supply for about $4.4 million.</strong></li>
 <li><strong>Only 7 out of more than 18,000 DAO members voted on the malicious proposal.</strong></li>
 <li><strong>BONK fell as much as 10% while Upbit and Kraken froze deposits and withdrawals.</strong></li>
</ul>
<p>On July 6, 2026, an unknown actor <strong><a href="https://arkm.com/explorer/address/CyEE7oHVDaFJ5xZLbXY3h2Z2uk1VwhTkdy72kPUEtypQ" rel="noopener noreferrer nofollow">transferred 4.43 trillion BONK tokens</a></strong>, worth roughly $20 million, out of the BonkDAO treasury and into a wallet under their control. There was no exploit, no compromised private key, and no bug in the smart contracts. The attacker bought slightly over 1% of the total BONK supply for around $4.4 million across Binance, Bybit and several DeFi lending markets, submitted a treasury transfer proposal through <strong><a href="https://v2.realms.today/dao/84pGFuy1Y27ApK67ApethaPvexeDWA66zNV8gm38TVeQ/proposal/6wR1jdhhJ31bbdRNXva8MxqsgsNLKTxargcdAyZ7FcRj/votes" rel="noopener noreferrer nofollow">Solana’s Realms platform</a></strong>, and voted it through. Out of more than 18,000 DAO members, exactly 7 wallets showed up to vote. Turnout came in at 2.9%, the proposal passed with 99.9% approval, and the funds moved the moment the vote closed.</p>
<p>BonkDAO and law enforcement are treating the incident as a criminal cyberattack. On-chain, every single transaction was valid. Both statements are true at the same time, and that is the actual story here.</p>
<h2 class="text-text-100 mt-3 -mb-1 text-[1.125rem] font-bold" id="strong-the-raid-cost-4-4-million-and-met-zero-resistance-strong"><strong>The raid cost $4.4 million and met zero resistance</strong></h2>
<p>The mechanics were closer to a hostile corporate takeover than a hack. The attacker accumulated voting power quietly, using centralized exchange purchases and borrowed tokens from lending markets to avoid moving the price. Once the position crossed the quorum threshold, they filed the proposal. Low participation did the rest. With 2.9% turnout, a holder controlling 1% of supply does not need to persuade anyone. They simply outvote an empty room.</p>
<p>After the treasury cleared, the attacker renamed the organization to “Sowellian BonkDAO” on Realms. <strong><a href="https://x.com/bonk_inu/status/2074191403781906800" rel="noopener noreferrer nofollow">BonkDAO confirmed the drain publicly</a></strong> about 20 hours later, stating it had identified the exchange wallets used to buy BONK ahead of the proposal and that it is working with exchanges, bridges, the Solana Foundation and law enforcement to trace the funds.</p>
<div>
 <table>
  <thead>
   <tr>
    <th>Metric</th>
    <th>Value</th>
   </tr>
  </thead>
  <tbody>
   <tr>
    <td>Cost of attack</td>
    <td>~$4.4 million</td>
   </tr>
   <tr>
    <td>Funds drained</td>
    <td>4.43 trillion BONK (~$20 million)</td>
   </tr>
   <tr>
    <td>Supply acquired</td>
    <td>Just over 1% of total BONK</td>
   </tr>
   <tr>
    <td>Wallets that voted</td>
    <td>7 of 18,000+ members</td>
   </tr>
   <tr>
    <td>Voter turnout</td>
    <td>2.9%</td>
   </tr>
   <tr>
    <td>Approval rate</td>
    <td>99.9% “Yes”</td>
   </tr>
  </tbody>
 </table>
</div>
<h2 class="text-text-100 mt-3 -mb-1 text-[1.125rem] font-bold" id="strong-why-the-treasury-moved-the-second-the-vote-ended-strong"><strong>Why the treasury moved the second the vote ended</strong></h2>
<p>Security analysts point to two missing safeguards that turned a bad vote into an irreversible loss. The first is the absence of an execution time-lock. In better-designed DAOs, a passed proposal sits in a mandatory waiting period of several days before funds actually move. That window gives the community time to spot a malicious transfer and evacuate the treasury. BonkDAO had no such delay. The payout executed automatically at the close of voting.</p>
<p>The second gap was the lack of an emergency multisig. Many protocols keep a council of trusted developers or independent researchers who can veto an obvious treasury drain, even after a vote passes. BonkDAO ran without one. The combination meant the entire security of a $20 million treasury rested on voter turnout, and turnout was 2.9%.</p>
<p>There is also a simple economic flaw underneath everything. When control of a treasury costs $4.4 million and the treasury holds $20 million, the attack pays for itself more than four times over. That imbalance existed on-chain, in public, for anyone to calculate.</p>
<h2 class="text-text-100 mt-3 -mb-1 text-[1.125rem] font-bold" id="strong-capital-has-beaten-community-before-strong"><strong>Capital has beaten community before</strong></h2>
<p>BonkDAO is not the first DAO to lose its treasury to a perfectly valid vote.</p>
<div>
 <table>
  <thead>
   <tr>
    <th>Protocol</th>
    <th>Year</th>
    <th>Method</th>
    <th>Outcome</th>
   </tr>
  </thead>
  <tbody>
   <tr>
    <td>Beanstalk Farms</td>
    <td>2022</td>
    <td>Flash loan to buy instant voting majority</td>
    <td>$182M drained</td>
   </tr>
   <tr>
    <td>Build Finance</td>
    <td>2022</td>
    <td>Token accumulation, quorum met on low turnout</td>
    <td>Mint rights seized, treasury emptied</td>
   </tr>
   <tr>
    <td>Tornado Cash</td>
    <td>2023</td>
    <td>Proposal <strong><a href="https://x.com/samczsun/status/1660012956632104960" rel="noopener noreferrer nofollow">granting attacker 1.2M fraudulent votes</a></strong></td>
    <td>Full control of DAO and parameters</td>
   </tr>
   <tr>
    <td>Compound</td>
    <td>2024</td>
    <td>Whale coalition accumulated COMP, passed proposal 682K to 633K votes</td>
    <td>$24M allocation approved, reversed after community pushback</td>
   </tr>
   <tr>
    <td>Balancer TOP pools</td>
    <td>2026</td>
    <td>Governance takeover of token pools</td>
    <td>$1.58M drained</td>
   </tr>
   <tr>
    <td>BonkDAO</td>
    <td>2026</td>
    <td>1% supply purchase, 2.9% turnout</td>
    <td>$20M drained</td>
   </tr>
  </tbody>
 </table>
</div>
<p>In five of the six cases the code executed exactly as written, and the losses came down to governance design rather than software failure. Compound is the partial exception, and it proves the turnout point from the other direction. When the Golden Boys coalition finally <strong><a href="https://compound.finance/governance/proposals/289" rel="noopener noreferrer nofollow">pushed its $24 million treasury proposal through in July 2024</a></strong>, on its third attempt, over 1.3 million votes showed up on both sides and the margin was under 50,000. The community contested the result immediately, and Compound negotiated <strong><a href="https://compound.finance/governance/proposals/289" rel="noopener noreferrer nofollow">the allocation’s cancellation</a></strong> in exchange for a new staking program, reportedly backed by a threat to strip the proposer’s voting power through a centralized multisig. Saving the treasury required abandoning the pretense that code alone governs. BonkDAO’s proposal faced 7 wallets.</p>
<h2 class="text-text-100 mt-3 -mb-1 text-[1.125rem] font-bold" id="strong-is-this-theft-or-a-legal-capitalist-maneuver-strong"><strong>Is this theft or a legal capitalist maneuver</strong></h2>
<p>The code-is-law argument holds that no exploit occurred. The accumulation, the proposal, the vote and the payout were all valid transactions recognized by the blockchain. If a DAO prices control of its treasury cheaper than the treasury’s contents, buying that control is an economic maneuver, not a crime. Proponents add a sharper point: a system cannot market itself as trustless and free of human intervention, then call the police when the mathematics work against it.</p>
<p>The opposing view treats rules as bound by social purpose, and it currently has the institutional weight behind it. BonkDAO, security firms and law enforcement have all classified the incident as bad-faith asset conversion, which most legal systems call theft. Acquiring governance power for the sole objective of turning 18,000 members’ shared assets into personal profit fits that definition regardless of what the blockchain recorded. Traditional corporate law already protects minority shareholders from raids by majority blockholders, and lawyers in this camp argue DAOs will need comparable human overrides if they want to operate globally. An active investigation is underway.</p>
<h2 class="text-text-100 mt-3 -mb-1 text-[1.125rem] font-bold" id="strong-bonk-price-slides-below-all-major-moving-averages-strong"><strong>BONK price slides below all major moving averages</strong></h2>
<p>The market reacted immediately. BONK dropped between 7% and 10% in the 24 hours after confirmation, while Upbit and Kraken <strong><a href="https://status.kraken.com/" rel="noopener noreferrer nofollow">suspended deposits and withdrawals</a></strong>to contain the risk of the attacker dumping the stolen tokens through their order books.</p>
<p><img src="https://ethnews.com/wp-content/uploads/2026/07/bonkusdt-chart-07-07-2026-tv.jpg" alt="BONKUSDT chart from TradingView - 07.07.2026. Shows RSI and moving averages (50,100, 200 SMA)" width="1280" height="740"></p>
<p>On the 4-hour chart, BONK trades at $0.00000422 after a sharp rejection from the early-July rally above $0.00000500. The price has fallen back below all three major simple moving averages, with the 50-period at $0.00000450, the 100-period at $0.00000440 and the 200-period at $0.00000445. A moving average simply shows the average price over a set number of candles, and when price sits below all of them, sellers control the trend across short, medium and long horizons at once. The three averages are also compressed within a narrow band just overhead, which now acts as a resistance zone around $0.00000440-$0.00000450.</p>
<p>The RSI, an indicator measuring whether recent moves have been dominated by buying or selling, reads 34.92. That is close to the 30 level traders consider oversold, meaning the selling has been aggressive but has not yet reached the point where rebounds historically become likely. The nearest meaningful support sits at the $0.00000400 area, which held the late-June lows. A confirmed break below it would open the chart toward the June bottom near $0.00000380.</p>
<h2 class="text-text-100 mt-3 -mb-1 text-[1.125rem] font-bold" id="strong-what-changes-from-here-strong"><strong>What changes from here</strong></h2>
<p>The immediate variable is the stolen supply itself. 4.43 trillion BONK in hostile hands is a standing liquidation threat, and every exchange that resumes withdrawals reopens a potential exit route. Traders will watch the attacker’s wallet for movement toward bridges or mixers, since any large transfer would likely trigger another leg down.</p>
<p>For the broader Solana DAO ecosystem, the incident lands as a direct stress test of Realms-based governance. Treasuries that lack time-locks and emergency councils are now visibly mispriced targets, and several protocols should be expected to push emergency parameter changes in the coming weeks. The larger unknown sits in court. Prosecutors have never had to attach criminal liability to on-chain actions that broke no code, and whatever precedent comes out of the BonkDAO case will decide how much legal protection a DAO treasury actually has beyond its own quorum math.</p>
<p>The post <a href="https://ethnews.com/bonkdao-20m-treasury-vote/" rel="noopener noreferrer nofollow">BonkDAO Lost $20M Because Only 7 People Showed Up to Vote</a> appeared first on <a href="https://ethnews.com" rel="noopener noreferrer nofollow">ETHNews</a>.</p>

BonkDAO Lost $20M Because Only 7 People Showed Up to Vote

Table of Contents

ETHNews

Other articles published on Jul 7, 2026

Join our free newsletter for daily crypto updates!