Rewards will be auto-burned anyway. And I am creating a Hive LLM, with every tool needed to perform the above. For now, I'll store it on IPFS until it is perfected, then start the transition to Hive. Only when the data is crisp, clean, in bulk, and redundancy-free will it ever touch the chain. I'll also check the chain for what info we already have and organize that. With HAF + Hive + AI, you have so much you can do in terms of data flexibility. Plus, people like me will have pretty much endless hard drive space, so I'll just be storing huge amounts of data, 100s of TB worth, personally myself, spending time distilling it all to its purest form. Right now, AIs scour the internet, searching various links, etc., which skips the middleman in the long run: all open source, all in one place, all Hive.