K
All
Discover
About
FAQ

About GM.xyz

What we're building

Our immediate goal is to make creating and managing a DAO or crypto-asset community as easy as spinning up a subreddit. Longer-term, we hope to build the first crypto-native, decentralized, and user-owned social network.

The problems we hope to address

1.

The tools currently used to create and manage crypto-asset communities and DAOs are complex and suboptimal. Discord—the de facto standard—is a chat app built for social gaming. Anyone who has joined a popular server and tried to sift through thousands of unordered messages knows there's much room for improvement. Today, communities have to stitch together a hodge-podge of products to accomplish their goals around token issuance, access gating, multisig management, community rewards, on-chain governance, and more. We believe this will have to be much easier if crypto-asset communities are going to be as big as we think they are.

2.

Traditional Web2 social networks (e.g. Reddit, Facebook, etc.) have take rates of 100% (see Chris Dixon Thread: "Your Take Rate is My Opportunity"). This should be closer to 1% or 2%. We believe community and content creators should capture almost all of the value for their contributions.

3.

Today's social media oligarchs have too much power and misaligned incentives. We believe a social network should be decentralized (nobody can control it) and permissionless (anyone can use it).

Progressive Decentralization

Today, our technology stack runs on Amazon Web Services, and it is pretty much identical to the stacks of most traditional Web2 applications. Our reasoning for this is that building a social platform people actually want to use is difficult enough with traditional architecture. Trying to build one today on decentralized architecture would be 100x harder and prohibitively expensive. Thus, our plan is to "progressively decentralize". This will involve open-sourcing our code base, opening up our APIs , and eventually moving to decentralized architecture (e.g. IPFS, Arweave, StarkWare, etc.).

After we have achieved product/market fit, formed a robust community, and put in place a model that properly incentivizes sustainable operations, we will "exit to the community". This model borrows heavily from the playbook outlined by Jesse Walden and Andreessen Horowitz here.