Why are almost everyone talking about AI on the blockchain, but few people clearly explain why AI actually needs blockchain technology?

If it's just about deploying models on the chain, that's only changing the runtime environment and doesn't solve any fundamental issues.
Recently, I've been looking into @0G_labs, and their approach is actually more straightforward. They don't start from the model itself but focus on the two most fundamental elements: data and computation. In simple terms, they are trying to build a blockchain-based infrastructure specifically for AI services—data and computing resources.
There's a key point that many people overlook: the core of AI isn't the model itself but the efficiency of data acquisition, storage, and retrieval.
What @0G_labs is doing is breaking these processes apart and handling them in a modular way, rather than stacking everything into a single system.
However, there's also obvious controversy: does this design really outperform existing cloud services in efficiency, or does it just add an extra layer of complexity?
If performance and cost advantages can't be achieved, decentralization alone is hard to justify.
But if it can truly establish advantages in data availability and computation scheduling, then it's not just solving an AI application problem but addressing the underlying supply of AI resources.
Many people are still wondering whether this is an AI project, but a more worthwhile question is whether they are rebuilding a layer of infrastructure.
@Galxe @GalxeQuest @easydotfunX @wallchain #Ad #Affiliate @TermMaxFi