Bittensor Patch #7: Network-Owned Validation Bittensor currently has two approaches to determining miner rewards Yuma combines scores from multiple validators through stake-weighted consensus NULL uses one validator’s scores, reducing duplicated validation work but concentrating trust in that validator I propose a third route A network-owned validation engine that runs each subnet’s scoring mechanism and submits the resulting weights Input: scoring rules, the data needed to apply them and miner outputs Output: weights Create a common interface so subnets can plug their scoring mechanisms into the same shared infrastructure Each subnet defines what good work looks like The engine handles the evaluation Fund it through emissions, limited to the verified cost of running the service It could purchase the resources it needs directly from subnets, such as compute from Targon and storage from Hippius Publish its spending so the network can see exactly what validation costs The engine would be owned and governed by the network, with open-source code and independently auditable execution so anyone can check that it works as intended For non-deterministic scoring mechanisms, subnets could specify multiple evaluations and how to aggregate the results Reward outsiders who uncover flaws in the validation engine, with clear rules for resolving challenges and correcting affected weights Keep upgrades and spending under network governance, with safeguards and failover to keep validation running reliably Subnets would still need well-designed scoring mechanisms But they could share the infrastructure that runs them, with emissions covering the resources and checks actually needed The result: • Auditable, network-owned validation • Open scrutiny • Validation at cost, without duplicated work • Demand for Bittensor’s own services Verify the scoring without relying on one validator’s word or paying many to repeat the same work $TAO

Bittensor Patch #6: Emission Taper Subnets should earn continued support as they mature Gradually reduce a subnet’s emissions over time, but let it slow or reverse that taper by returning value to alpha holders through revenue-funded buybacks and burns or direct distributions The more value a subnet gives back, the more continued funding it can justify Those that don’t would see their emissions gradually decline Measuring that returned value in TAO terms lets emissions adjust dynamically as TAO’s price changes As TAO’s price rises, the same dollar buyback represents less TAO and therefore does less to slow or reverse the taper This helps prevent the network from simply paying more for the same work because $TAO became more valuable To sustain a larger dollar subsidy, a subnet should deliver more value back Combined with the previous patch, subnets could choose how quickly they consume their subsidy, with lower emission rates also slowing the taper That preserves room for long-term research while raising expectations for mature subnets Full article ↓ xitter.cf/BittensorIntern/status…