Did Arbitrum just commit the first DAO seppuku?
The 2nd biggest TVL layer two network, Arbitrum DAO, airdropped 1.275 billion tokens a week ago. After the drop, the price of ARB has not tanked and the number of tokenholders is actually rising to nearly 300k wallets. There is even a Snapshot proposal aiming to establish the Constitution of the Arbitrum DAO. What could possibly go wrong?
Apparently, somebody in the Arbitrum Foundation forgot how time works. While being phrased as a request for the future allocation of funds, the proposal (AIP-1) was apparently meant as a rubber stamp to ratify the spending already happening.
It appeared that the proposal advocated for allocating to the Foundation $750M out of the $4.27B previously allocated to the DAO treasury without being especially clear about where that money would go.
Furthermore, keen observers noted that wallets belonging to Arbitrum core team already moved some of that $750M, converting some to stablecoins and spending it on various costs of running Arbitrum.
Understandably, a number of ARB holders and delegates became outraged at this spending of funds yet to be approved by the DAO. Some top delegates voted NO on the proposal and it ultimately failed with nearly 77% of voters going against it. Here is the detailed reasoning of Arbitrum’s 2nd biggest delegate for his NO vote and his take on what happened.
The team then backtracked and explained that only $3.52B was meant for the treasury and the $750M was always meant for the Foundation. Yet nowhere in its tokenomics was this previously mentioned.
Note that it’s not controversial for the Foundation to control a significant share of the governance token. Arbitrum’s allocation equals 7.5%, which is in the middle of how much Starknet, Optimism, and Polygon kept from their own drops.
Rather, what upset people so much is the convoluted (or fraudulent, according to some) way in which Arbitrum’s foundation declared this post factum and tried to pass it off as a genuine proposal up to the DAO’s members to decide. Pushing proposals on something that is already going on regardless of the vote makes a mockery out of the very structure of DAOs.
So what’s next? The vote concludes tomorrow morning and does not look promising for the Foundation. At the same time, there is no indication that the top delegates and a majority of the voters are against funding the Foundation with as much ARB as it needs. There does not seem to be a mass exodus from ARB either in holdings or holders.
What has happened is that the trust in the founding teams of DAOs has been shaken heavily, especially since Arbitrum’s team was previously generally well-trusted. With Arbitrum’s sister network, Arbitrum Nova, potentially spinning out its own DAO airdrop, it will certainly face greater scrutiny.
On the positive side, this SNAFU did show the effectiveness of DAOs in letting any member use the public blockchain to discover any alarming transactions and sound the alarm bells. Better yet, it showed the power of DAO members to stop potential abuse from a centralized group of people, proving that DAOs do work!
Another thing to note is the need for a clear mechanism to modify a DAO once created. For example, if the next “Arbitrum” uses DeXe’s DAO constructor, omitting the allocation to the Foundation could be fixed by making the appropriate proposal to grant the Foundation a certain amount of funds. There could even be proposals to set up special rules for big decisions like allocating funds to the core team with, for example, more stringent quorum settings to encourage a wide community consensus. Because, as Arbitrum DAO members have shown — governance token holders are smart and don’t like to be misled.
Stay tuned!
Website | Telegram channel | Telegram chat | Facebook | Medium| LinkedIn | Twitter | Reddit| Discord