TAO First Halving: What You Should Know

Author: Lucas Source: X, @OnchainLu Translation: Shan Ouba, Golden Finance

The crypto world is about to witness another "halving", but this time the protagonist is not Bitcoin, but Bittensor — a decentralized AI network known as the "smart Bitcoin". Its first halving event is expected to occur between December 2025 and February 2026.

Similar to Bitcoin, TAO has a total supply of 21 million coins, with inflation controlled by periodic halving; however, Bittensor's incentive structure spans multiple subnets, each with its own native token Alpha (denoted as 'α') and independent liquidity mechanism.

The complexity of this structure makes the "chain reaction" brought about by this halving more difficult to predict.

The issue is not only whether the price of TAO will rise, but whether the entire multi-subnet ecosystem can withstand: less liquidity injection after the halving, the dilution exacerbation of Alpha relative to TAO, and the potential sustained selling pressure from cross-subnet Alpha.

The purpose of this article is to clarify the operational mechanism of TAO halving and predict the potential impacts on different participants in the ecosystem.

Known Facts

Let's first take a look at the known information.

When the circulating supply exceeds 10.5M TAO (exactly 50% of the total cap of 21M), a halving will occur. As of the time of writing, the network currently has approximately 9.4M TAO in circulation, and Taostats' live timeline estimates that this threshold will be reached on December 13, 2025.

Since the halving is based on "circulation" rather than block height, its exact timing will be affected by the TAO recovery rate. Actions such as miner/validator deregistration, subnet registration fees, cold key replacements, etc. will recover TAO back to the unissued pool, thereby delaying the halving time. If these activities decrease and TAO recovery diminishes, the halving will come earlier.

Contrary to what you might think, the TAO halving is not a halving of participant rewards.

When the halving occurs, the following will happen:

Block rewards and issuance halved.

• Each block will mint 0.5 TAO instead of 1 TAO. Considering that a block is generated approximately every 12 seconds, this means that about 3,600 TAO will be issued daily (after halving), instead of about 7,200 TAO daily (before halving).

Network inflation decreases.

• TAO inflation will decrease from approximately 25% per year to about 12.5% per year (based on a base supply of 10.5 million, although the actual increase in supply will be lower, as some tokens will be burned).

Alpha_in (Halving of Alpha injected into each subnet transaction pool).

• This is because Alpha_in is calculated by dividing the TAO issuance by the Alpha price, and the TAO issuance has been halved. As Alpha_in is responsible for balancing liquidity, its halving will make the pool shallower, meaning that any transaction—whether a buy or sell—will result in greater price fluctuations. This effect will be more pronounced in subnetworks with lower liquidity before the halving.

Alpha_out (the Alpha rewarded to miners/validators/owners) remains unchanged.

• This means that the issuance will remain unchanged. Alpha_out will only halve during the individual halving of each subnet, following a separate schedule.

Downward price pressure on Subnet Alpha has increased.

• With Alpha_in halving (at least initially) while Alpha_out remains unchanged, there will be stronger selling pressure, especially in subnet trading pools with shallow liquidity. To clarify, "stronger selling pressure" does not necessarily mean more people are selling—what I mean is that due to the lower liquidity of Alpha in the subnet trading pool, the same amount of Alpha sold will have a greater impact and slippage on the price.

Root APY moderately decreased.

• As Root rewards are distributed in TAO, the increased slippage experienced during Alpha conversions (especially in low liquidity pools) will gradually reduce the dollar value of the rewards distributed to Root stakers. This does not take into account broader market dynamics.

Unless upgraded separately through governance, the block time, burn/recovery, and subnet reward distribution remain unchanged.

What does that mean?

Now let's talk about how these structural changes will affect various participants in the Bittensor ecosystem. If you are not familiar with the roles of subnet owners, miners, or validators, you can refer to this documentation.

Subnet Owner

In the context of halving liquidity injections, building a successful subnet will become exponentially more difficult.

All subnets will be affected after a period of time, but the impact on new subnets is the most direct, as their trading pools have not yet accumulated sufficient liquidity. For example, a subnet that was launched six months ago may already have $100,000 of TAO supporting its trading pool; while a new subnet launched after the halving may struggle to reach even $50,000 in the same time frame.

Due to the halving of TAO available for trading, each buy and sell order will have a greater impact on the Alpha price. The previously mentioned downward price pressure will have a cascading effect on Alpha, especially for new subnetworks that do not have reserves in the pool, making it very difficult to bear.

In order to alleviate this pressure as soon as possible, I believe that new subnet owners need to:

• Find product-market fit (PMF) within weeks, rather than months, or miners and investors will be forced to migrate to more liquid alternatives. The time window will become more urgent.
• Place greater emphasis on the creation of external income, for example:

• Charge customers for inference services (e.g., Targon),
• Sales and authorization data (e.g., Data Universe, Masa),
• or other forms of application fees.

Currently, there are indeed a small number of subnet owners operating like startup CEOs, with real customers and sources of income, rather than relying on yield farming. The arrival of the halving will accelerate the urgency for everyone to shift to this model.

Miners and Validators

The impact on miners and validators is more complex -- their Alpha rewards (Alpha_out) and operational costs (such as electricity, hardware costs, and maintenance fees) will not change, but due to the decline in liquidity supported by the transaction pool of each subnet, their risk of earnings has significantly increased. In a situation where profit margins are compressed, the price performance of Alpha will determine whether miners and validators can be profitable. If Alpha maintains a stable price, they can still make a profit in the subnet; but if Alpha plummets due to increased volatility or selling pressure, operational costs may consume all returns, leading to losses. I expect that most miners and validators will withdraw from those subnets with high volatility and low liquidity, focusing instead on those subnets with strong TAO support and clear business models.

We will see a shift from "quantity first" to "quality first" — in those liquidity-rich, verified subnetworks, UID (subnetwork identity) will become scarce. As UID competition intensifies, miners will no longer be able to "cast a wide net" and invest minimal effort across multiple subnetworks to achieve substantial returns. They will be forced to focus on 1-2 high-quality subnetworks and deliver truly valuable computational work in order to obtain sustained earnings. After the halving, the key to success will be specialization and stable high-quality contributions to compete for the limited Alpha rewards.

In comparison, the adaptation difficulty for validators is relatively low, as most validators run validation scripts provided by subnets, requiring little specialized skills or operational changes. The main challenge they face is: how to choose the right subnet. The main rewards for validators are still determined by staking weight (their Alpha holdings, plus 18% of TAO held or delegated), and they can still maintain a diversified strategy across multiple subnets and reallocate their staking funds to the most robust opportunities.

Ultimately, both miners and validators will face a "winner-takes-all" situation: participants will concentrate computing power and staking on those subnets with the strongest liquidity and best performance. This may result in the rewards of these popular subnets being divided among more people, even in the "winner" subnet, the unit returns for each participant may decrease.

What does that mean for investors?

Now, we finally arrive at the part you might be looking forward to the most.

Staking Dynamics: Funds "Fleeing to Safe Havens"

The halving will trigger a large-scale reallocation of capital, as stakers will face uneven yield compression throughout the ecosystem. We can already see early signs of this shift.

From the above figure, the amount of TAO staked on Alpha has been steadily increasing, but the pace of this growth has started to slow down. The recent deceleration over the past month does not mean that more people prefer staking Alpha over Root — it simply indicates that new TAO has entered circulation, with a portion allocated to dTAO (subnet).

The TAO staked on Root (as shown in the figure) exhibits an interesting trend.

As investors become more familiar with dTAO staking, and under the pressure of declining Root annualized yield (as shown in the figure below, currently around 8–10%), Root stakers have gradually reduced their holdings. Although there has been a noticeable rebound in Root staking volume since the end of June, rising from about 5.7 million to 6.1 million, this surge mainly stems from a technical correction — approximately 400,000 TAO that were originally on-chain existed in wallets but were not correctly accounted for in the total staking volume.

The key question is: will this upward trend of Root staking continue after the halving?

I think so, and it will be significantly sustained.

As we have discussed, most Alpha tokens will become more volatile and carry higher risks after the halving. I don't believe there are many rational investors who would still be willing to accept the same Alpha yield in the face of greater downside risk.

Although Root's annualized yield will be somewhat affected by increased slippage, it offers stability with TAO exposure, and will not fluctuate as wildly as the illiquid Alpha pool. I believe that directly holding TAO through Root staking will become a more risk-adjusted advantageous choice. As for those investors who still choose to continue staking subnets instead of Root, I expect they will focus on those subnets that have established liquidity, a clear revenue model, and real use cases.

Conclusion

This TAO halving will create a screening mechanism, which could be favorable for TAO itself in the long term. The liquidity injection being halved will weaken the safety buffer on which weak subnets that rely solely on emission for survival depend, forcing them to shift from "relying on emissions" to "real revenue generation."

This stress test is likely to eliminate inefficient subnets while also forcing the survivors to further amplify their advantages. Those strong subnets that can successfully generate external revenue will become the only sustainable destination for miners, validators, and stakers.