The “Root Reborn” proposal would enable TAO validators to allocate capital across selected subnets and reinvest rewards, rather than continuously selling subnet tokens to fund staking payouts. The proposal is still under review and has not yet been implemented.
A new governance proposal for Bittensor, the decentralized AI network behind the TAO token, aims to reshape validator incentives by positioning them more like active fund managers.
Bittensor is composed of numerous subnets, each serving as a marketplace for specific AI tasks and operating with its own token. TAO functions as the network’s primary asset, with users earning yield by staking it to validators on the root layer, considered the network’s most secure capital base.
The proposal introduces a major shift in how staking rewards are generated and distributed.
Currently, rewards are funded by selling subnet tokens earned by validators and converting them into TAO for stakers. This mechanism creates persistent sell pressure on subnet tokens, weighing on their valuations.
The “Root Reborn” proposal, introduced by developer “unconst,” seeks to reverse that dynamic.
Rather than liquidating rewards, validators would build portfolios of selected subnets. Rewards would be reinvested into these allocations, forming a compounding basket that is restaked with the validator. Stakers would still receive yield and retain the option to convert it into TAO at any time.
This model could eliminate ongoing sell pressure and instead introduce net buying, potentially supporting subnet token prices.
Validators would transition from passive yield distributors into active capital allocators. Subnets they back could attract more capital, while weaker or less credible projects may see reduced funding.
The proposal has been submitted to Bittensor’s GitHub and is currently targeted for testing on a testnet, rather than immediate deployment on the main network.
An initial automated review flagged two key risks: a potential upgrade bottleneck involving large data volumes and a payout issue that could disadvantage stakers if a subnet shuts down. The author has since stated that these issues have been addressed, with further refinements planned ahead of any mainnet release.
Over the past year, TAO has declined roughly 28%, compared with a 38% drop in bitcoin over the same period. Meanwhile, staking TAO currently offers an annual yield of about 17% for long-term holders.
