Johnny@empyrealdev
I wanted to take the opportunity to explain what makes Simulacrum different from a lot of projects that have been popping up that we're getting compared to. Most of the ones I've seen have minimal documentation on how they work, but we can safely infer that most of them are centralized executors running your tweets through an LLM to translate them into transactions that they execute on your behalf.
Simulacrum, on the other hand, is a framework for adding account abstraction to arbitrary public identity based social networks. For example, you can add account abstraction to X. Tweets are indexed, attestations are gathered for the tweet, and they are submitted to an onchain synthetic blockchain to be assembled into blocks for execution.
This is in sharp contrast with a centralized executor that runs your request through an LLM and translates it to a transaction. This would lack self-custody and introduces the same problem we saw with a lot of the telegram bots, where it ends up being a team controlling a bunch of wallets and you're blindly trusting them to do the right thing.
With Simulacrum, no command can be executed unless it was tweeted. Likewise, anyone can submit a command onchain to be indexed once they have gotten attestations to its validity. To compare us to a custodial solution where a bot manages your wallet for you is a false comparison.
What does this all mean? Simulacrum is not a centralized, trusted executor. It is a synthetic blockchain that enables trustless execution of offchain intents submitted via arbitrary, publicly verifiable sources. This is important. Centralized companies can be compromised, they can be closed. Decentralization is a security feature, and an important one at that.
P.S. We will add natural language based execution, but we will be doing this in a way where it's not just a centralized executor. We wanted to give ample time for people to familiarize with the core concept before we introduced natural language. As many of you know, it's very easy to convert unstructured data to structured data using a language model. So establishing a structured command format is essential to making a platform that plays nicely with natural language. The groundwork we have done was essential to maintain the trustless nature of the platform as we continue to enhance its capabilities.
