Dynamic TAO: Bittensor’s new economic model

Author: @OnchainLu

Compiled by: Felix, PANews

If you are reading this article, you are already familiar with the Bittensor ecosystem and are aware of the Dynamic TAO upgrade on February 13th. If not, it is recommended to read Dynamic TAO for Dummies , which describes the high-level changes and impacts of dTAO on the Bittensor ecosystem. The purpose of this article is to deeply analyze how the introduction of dTAO will change the issuance distribution, incentive structure, and economics of the network.

Issuance Mechanism

dTAO brings a fundamental shift to Bittensor’s economic model by implementing subnet-specific Alpha tokens that trade against TAO on a constant product automated market maker (AMM) - here’s an explanation of how the AMM works with dTAO. With this new mechanism, the relative price of a subnet Alpha token directly affects the amount of TAO issuance it receives, rather than a small group of validators controlling issuance allocation.

Release Components

The new release will consist of three parts:

• TAO Allocation Based on Subnet Alpha Token Price (Part 1)
• Injecting Alpha into Subnet Liquidity Pools (Part 2)
• Additional Alpha Distribution Distributed among Subnet Owners, Validators, and Miners (Part 3)

These issuances are calculated every block (approximately 12 seconds).

TAO issuance formula

The core TAO issuance (part 1) formula is as follows:

Subnet TAO issuance = (Subnet Alpha price / Total of all subnet Alpha prices) × (Total TAO issuance per block)

in:

• Subnet TAO Issued → The amount of TAO issued to a specific subnet
• SubnetAlphaPrice → Price of SubnetAlpha token
• All subnet Alpha price sum → All subnet Alpha token price sum
• Total TAO issuance per block → Total TAO issuance per block (1 TAO)

This formula distributes TAO issuance based on the relative market value of each subnet’s Alpha token. Subnets with higher demand and liquidity will receive a larger share of TAO issuance, thereby incentivizing valuable services and user appeal.

Alpha Token Injection (Part 2)

Alpha injection follows a similar but modified formula:

Alpha injection = min( [Total TAO issuance per block / Sum of all subnet Alpha prices] , [Subnet Alpha issuance cap] )

Key points:

• The total TAO issuance per block is initially 1 TAO, but follows a halving schedule.
• Subnet Alpha Issuance Cap → The maximum Alpha that can be injected into the subnet liquidity pool per block (initially 1 Alpha per block, also following the halving schedule).
• Alpha injection is proportional to the injected TAO, divided by the sum of all subnet prices (also capped by [Subnet Alpha Issuance Cap])

This mechanism provides liquidity to subnet AMMs while preventing excessive inflation.

Additional Alpha Release (Part 3)

In addition to the Alpha injected into the liquidity pool, there is additional Alpha issuance distributed to subnet owners, validators, and miners. Each subnet can issue up to 1 Alpha per block and follows the TAO halving schedule.

Allocation details:

• 18%: Subnet owners
• 41%: Validators
• 41%: Miners

This reward mechanism incentivizes subnet owners, validators, and miners to contribute to subnet operations, security, and growth.

Total Alpha issuance per block (before halving):

• Up to 1 Alpha is injected into the subnet liquidity pool (Part 2)
• Up to 1 Alpha distributed to subnet owners, validators and miners (Part 3)

It is important to note that both forms of Alpha issuance—Alpha injected into the liquidity pool (Alpha-in) and Alpha distributed to subnet participants (Alpha-out)—follow the same halving schedule as TAO.

To be clear, each subnet follows its own halving schedule. Subnets that launch earlier will experience periods of higher issuance rates because they launch at the beginning of the halving schedule. Subnets that launch later must accept the current (lower) issuance rate when they launch because all subnets follow the same halving thresholds based on predetermined supply milestones.

Synchronized halving of all issuance helps maintain predictable token supply growth and controls inflation across the system.

Example Calculation

Assume there are three subnets, with Alpha prices of 2 TAO, 1 TAO, and 1 TAO (4 TAO in total).

For TAO issuance of blocks (Part 1):

• First subnet → 0.5 TAO (2/4 × 1 TAO)
• The other two subnets → 0.25 TAO each (1/4 × 1 TAO)

Now let's focus on the corresponding Alpha injection (part 2):

Each subnet's Alpha issuance is capped at 1 per block (assuming they are all in the initial stages of their halving schedule), so they get min{0.25, 1}, min{0.25, 1}, and min{0.25, 1} Alpha respectively.

In addition to these pool injections, each subnet will receive an additional 1 Alpha (part 3), which is distributed between owners, validators, and miners in a ratio of 18/41/41.

This creates a powerful dynamic where higher value subnets naturally attract more TAO issuance, while the combination of injection caps and fixed rewards maintains economic stability. Subnets representing 50% of Alpha’s total market value will receive 50% of TAO issuance, creating a direct link between market value and resource allocation.

Alpha Price Manipulation?

You may be concerned about Alpha price manipulation. As the transaction size grows relative to the liquidity of the subnet pool, the slippage cost increases, and the constant product AMM can create a defense mechanism for Alpha price manipulation.

Example:

Assume that the subnet’s Alpha/TAO pool has 100,000 Alpha and 50,000 TAO → Alpha price = 0.5 TAO.

Purchasing 10,000 Alphas will cost 5,556 TAO, making the effective price per Alpha 0.5556 TAO (11% price impact).

In short, a trade that accounts for 1% of the pool liquidity will have about a 1% price impact, but a trade that accounts for 10% of the pool liquidity will have about an 11% price impact.

This makes large-scale manipulative trading extremely expensive while maintaining the efficiency of normal market operations.

Order sorting is determined randomly

There is one more thing to note for potential subnet investors: Bittensor uses a random order finalization mechanism, which means that the order of transactions executed in each block is not first-come, first-served.

For example, if many investors try to enter the same subnet liquidity pool within the same block (perhaps due to a coordinated effort or a team of copy traders), the order of their orders will be randomized, meaning:

The risk of price manipulation is reduced because attackers cannot effectively front-run orders.

The price impact and slippage experienced by each investor will be different, depending on the random order in which they trade within that block. Due to randomization, some may receive a better price than others.

While this mechanism is effective in preventing manipulation, it introduces unpredictability to large collaborative investments. Therefore, investors who wish to collaboratively invest in a particular subnet should be prepared that some members may face higher slippage costs than others. By reducing the predictability of the outcomes of transactions on the same block, it will ultimately encourage more organic and decentralized market dynamics.

Related reading: A comprehensive analysis of the Bittensor ecosystem: The Game of Thrones for the AI Iron Throne