The compute and training layers of the AI-industrial-complex are currently dominated by centralized Web2 giants that have unparalleled access to capital, state-of-the-art hardware, and vast datasets. While this will likely remain the case for the most powerful general ML models, mid-tier and bespoke models may increasingly source their compute resources from more affordable and accessible Web3 networks. Similarly, for inference needs that exceed the capabilities of personal edge devices, some consumers may turn to Web3 networks for less censored, more diverse outputs. Rather than attempting to overhaul the entire AI stack, Web3 challengers should focus on these niche use cases, and lean more heavily into their unique value proposition around censorship-resistance, transparency, and social verifiability.

One way to think of the design choices within the layers is the lower down the stack we go, the more we need small machines to be able to verify the truth. By ‘small machines’ we mean a set of requirements that are widely achievable, allowing nearly anyone to verify the truth if they so desire, keeping collusion at bay. So long as these smaller machines are publishing and verifying “the truth,” then bigger machines can run execution at levels higher up the stack, achieving greater performance but not sacrificing decentralization where it matters. Intriguingly, this design means Eclipse’s transaction throughput capacity could surpass that of the Solana L1, allowing it to be home to a diverse ecosystem of its own.