Bittensor (TAO) Bearish Logic: The Income Desert Beneath the Computing Power Myth

marsbitPublicado a 2026-03-24Actualizado a 2026-03-24

Resumen

Bittensor (TAO), currently priced around $275 with a market cap of $2.6 billion, faces significant challenges in generating sustainable external revenue despite its strong narrative around token scarcity, institutional backing, and AI decentralization. The network relies heavily on inflationary token emissions to subsidize operations, with an estimated $36 million in annual incentives dwarfing confirmed external revenues of only $3–15 million. Key subnet Chutes (SN64), for example, receives subsidies 22–40 times higher than its actual revenue. Without subsidies, its costs would exceed those of centralized competitors. The network lacks transparent demand-side metrics, and its subnets operate in a highly competitive environment, squeezed between low-cost self-hosted solutions and heavily subsidized cloud giants. While TAO’s value is driven by speculative factors like scarcity, halving events, and ETF prospects, its fundamental economic model shows little evidence of long-term viability or competitive advantage in the AI services market.

Author: Pine Analytics

Compiled by: Saoirse, Foresight News

TAO is currently priced around $275, with a market cap of $2.6 billion and a fully diluted valuation of $5.8 billion. The project has received institutional backing from Grayscale (which filed an application for a NYSE ETF listing in December 2025) and public recognition from NVIDIA CEO Jensen Huang. Its token supply narrative is also highly compelling: a hard cap of 21 million tokens with a Bitcoin-style halving mechanism. After the first halving in December 2025, the daily issuance will drop from 7200 tokens to 3600 tokens. Within a year, the number of subnets has grown from 32 to 128, and Templar's Covenant-72B training has proven that decentralized computing power can produce large language models with baseline competitiveness.

This report does not deny the above facts. What we aim to explore is: can the network's economic model generate real external revenue sufficient to support its current valuation, and what is its true competitiveness when pitted against centralized service providers and self-hosted computing power?

Bittensor (TAO) Token Issuance Allocation Ratio

How Network Value Flows

Bittensor has four types of participants:

  • Subnet owners build specialized AI markets and receive 18% of the TAO issuance rewards for their subnet;
  • Miners perform AI tasks (inference, training, data processing) and receive 41%, totaling about 1476 TAO daily, with an annualized value of approximately $148 million;
  • Validators score the output of miners and receive 41%;
  • Stakers deposit TAO into a subnet's liquidity pool in exchange for the subnet's native token.

Under the Taoflow model, a subnet's reward share is determined by the net inflow of TAO staked; negative net inflow results in no rewards. The top ten subnets control about 56% of the total network issuance.

TAO is the universal network token: miner registration, validator staking, subnet token purchases, and service payments all require TAO. Theoretically, subnet activity should create structural demand for the underlying token.

Comparative analysis of inference costs between Bittensor subnet Chutes (SN64) and centralized service provider LLaMA 70B model

Demand-Side Status Quo

Transparent Supply vs. Opaque Demand

Bittensor's supply side is highly transparent: 3600 TAO are distributed programmatically each day, the halving rules are hard-coded, and staking rates (~70%), allocation ratios, and flow data are all on-chain.

However, the demand side is completely opaque. There is no unified dashboard tracking external revenue per subnet. Actual AI service usage (inference, computation, training) occurs off-chain and is not recorded on the blockchain. Investors can only infer demand through indirect indicators like staking flow, subnet token prices, and self-reported data from projects. This opacity is structural, not temporary. The blockchain only records token transfers, not API calls.

Here is the most complete picture of the demand side as of March 2026.

Chutes (SN64): Low Prices Rely Entirely on Subsidies

Chutes commands 14.4% of the total network issuance, the highest of any subnet. Developed by Rayon Labs, it offers serverless inference services for open-source models, quoting prices 85% lower than AWS and 10%–50% lower than Together AI. Its usage data is unparalleled within the ecosystem: over 400,000 users (over 100,000 API users), over 5 million daily requests, cumulative processing of 9.1 trillion tokens, with a three-day average token generation surging from 6.6 billion to 101 billion. It is also a top inference provider on OpenRouter, with some models outperforming centralized competitors.

However, these low prices do not come from operational efficiency but from subsidies.

Calculated based on its 14.4% share, Chutes receives approximately 518 TAO daily, with an annualized value of about $52 million. Its external annual revenue is only about $1.3–2.4 million (the higher figure is self-reported by the team, not independently audited). The protocol's subsidy ratio for this subnet is approximately 22:1 to 40:1. For every $1 users pay, the network must release $22–40 worth of TAO through inflation as a subsidy.

Without subsidies, based on its daily processing volume of about 101 billion tokens, the cost price would be approximately $1.41 per million tokens. The current centralized market prices are:

  • Together.ai's LLaMA 3.3 70B Turbo ~$0.88 / million tokens;
  • DeepSeek V3 ~$0.40–0.80;
  • Small models can be as low as $0.18.

This means that without subsidies, Chutes' price would be 1.6–3.5 times more expensive than centralized solutions. The so-called 85% cost advantage is completely reversed; its low price is essentially paid for by TAO holders through inflation, not structural efficiency brought by decentralization.

When the next halving arrives (expected late 2026 or 2027), either prices will double, miners will leave, or the gap between subsidies and revenue will widen further.

Some might draw parallels to early internet subsidies for customer acquisition, but Uber, DoorDash, and AWS built switching costs during their subsidy periods: proprietary platforms, driver networks, enterprise ecosystems. Bittensor subnets have no barriers: models are open-source, interfaces are standardized, and users can switch providers at zero cost. Once subsidies recede, there is no lock-in mechanism to retain users.

Rayon Labs also operates SN56 and SN19, collectively controlling about 23.7% of the total network issuance. Neither has disclosed external revenue. A single team controls almost a quarter of the network's incentive distribution.

Targon, Templar, and Other Subnets

Targon (SN4) is the highest-revenue subnet, operated by Manifold Labs, providing confidential GPU computing services to enterprises. Estimated annual revenue is approximately $10.4 million, corresponding to a valuation of $48 million, a P/S ratio of about 4.6x, making it the most solid valuation within the ecosystem. However, the $10.4 million is a predicted figure cited by multiple reports, not an audited number.

Templar (SN3) completed the Covenant-72B training, with a market cap of $98 million, but has zero external revenue. Training API and enterprise sales are still in progress, with no paid product launched yet.

The remaining 120+ subnets either have no public revenue or are still in the early product stages, surviving primarily on token issuance subsidies.

Overall Picture

The total confirmed demand-side annual revenue for the entire network is only about $3–15 million. The annualized subsidy for Chutes alone (~$52 million) exceeds the upper limit of the entire network's external revenue.

Based on a $2.6 billion market cap, its revenue multiple is about 175–200x; based on the $5.8 billion fully diluted valuation, it's nearly 400x. In contrast, centralized AI computing companies have recently raised funds at valuations of 15–25x forward revenue, and high-growth SaaS rarely sustains above 50x for long. Bittensor's valuation multiple is 4–10 times that of aggressive industry benchmarks.

The huge gap between valuation and demand fundamentals indicates that the price of TAO is almost entirely based on supply-side scarcity (halving, staking lock-up), institutional catalysts (Grayscale ETF, exchange listing expectations), and AI sector sentiment, rather than real economic output. These are indeed price drivers, but they are completely separate from the logic of "Bittensor creating sustainable value as an AI service network."

Comparison of hyperscale cloud provider AI capital expenditure vs. Bittensor (TAO) annual subsidy scale

Pricing Dilemma: Squeezed from Both Sides

Subnets face pressure from two sides:

  • Above: Self-Hosting Ceiling

All models on the platform are open-source, weights are public, the comprehensive cost to run a 70B model on a single H100 is only $40–50 per day, and tools like vLLM and Ollama make local deployment extremely simple. NVIDIA's new generation of chips will further significantly reduce inference costs. Institutions with sufficient volume will find self-building deployment cheaper.

  • Below: Pressure from Cloud Giants

Microsoft, Google, Amazon, and Meta had combined AI capital expenditures exceeding $200 billion in 2025. They have hardware priority allocation, dedicated data centers, enterprise customer relationships, and can subsidize AI with cash flow from other businesses. Bittensor's annual incentive budget (~$360 million) is less than Microsoft's weekly AI infrastructure investment. Professional service providers also use VC subsidies to compete on price with open-source models.

Subnet pricing is compressed into an extremely narrow band, while also bearing decentralization-specific costs: token friction, validator node overhead, subnet owner share, network latency, etc.

The Moat Problem

Even if a subnet creates a valuable service, the underlying model and methods are inherently public: Covenant-72B uses the Apache license, and technical papers are published. Any competitor can replicate it directly without participating in the TAO ecosystem.

Traditional moats (proprietary technology, network effects, switching costs, brand) do not hold:

  • Technology is open-source;
  • Network effects belong to TAO, not individual subnets;
  • Model weights are identical, user switching cost is zero.

The community believes the incentive mechanism is the moat, but this relies on continuous large-scale token issuance, and each halving will steadily shrink the incentive budget.

What is TAO Actually Trading On

At a $2.6 billion market cap, TAO's price does not reflect demand fundamentals; $3–15 million in annual revenue cannot support it under any traditional framework. The market is trading on: Bitcoin-style scarcity, Grayscale ETF expectations, AI sector rotation, and the long-term option value of decentralized AI. These are all reasonable speculative factors, but they come entirely from the supply side and market sentiment.

If you hold TAO based on scarcity and narrative, you might profit even with weak demand; but if you believe Bittensor will become a truly large-scale AI service network, there is currently no evidence, and it encounters structural resistance that is difficult to break through. Investors should clearly distinguish their investment logic.

Preguntas relacionadas

QWhat is the main bearish argument against Bittensor (TAO) presented in the article?

AThe main bearish argument is that Bittensor's current valuation is not supported by its real external revenue. The network's demand-side revenue is minimal (estimated at $3-15 million annually), while its largest subnet, Chutes, relies on a massive 22:1 to 40:1 subsidy from TAO token inflation to offer competitive pricing. Without these subsidies, its services would be more expensive than centralized alternatives, and the network lacks a sustainable economic model or competitive moat to justify its multi-billion dollar valuation.

QHow does the subsidy model for the subnet Chutes (SN64) work and what does it reveal?

AChutes receives approximately 518 TAO daily (14.4% of the network's issuance), which is an annualized subsidy worth about $52 million. However, its external annual revenue is only between $1.3-$2.4 million. This creates a subsidy-to-revenue ratio of 22:1 to 40:1, meaning the network inflates the token supply by $22-$40 for every $1 of real revenue earned. This reveals that its low prices are artificial and dependent on unsustainable token emissions, not superior efficiency.

QWhat are the key challenges Bittensor faces in competing with centralized AI service providers?

ABittensor faces a pricing squeeze from two sides. From above, self-hosting is becoming cheaper and easier with tools like vLLM and Ollama. From below, tech giants like Microsoft, Google, and Amazon have massive AI capital expenditures (over $200 billion combined in 2025), hardware advantages, and the ability to subsidize services with other business revenues. Bittensor's entire annual incentive budget (~$360 million) is dwarfed by these competitors, and its subnets must also bear the additional costs of decentralization, such as token friction and network latency.

QWhy does the article claim that Bittensor subnets lack a traditional economic moat?

AThe article claims subnets lack a moat because their core technologies are open-source (e.g., Covenant-72B uses an Apache license), which allows any competitor to replicate their models and services without participating in the TAO ecosystem. There is no proprietary technology, no significant network effects that benefit individual subnets (the effects belong to TAO itself), and zero switching costs for users since model weights are consistent across providers. The only perceived moat, the incentive system, relies on continuous and substantial token emissions, which are scheduled to halve.

QWhat does the article conclude is the primary driver of TAO's current market price and valuation?

AThe article concludes that TAO's price is primarily driven by speculative factors on the supply side and market sentiment, not fundamental demand or revenue. These factors include its Bitcoin-like scarcity narrative (fixed supply and halvings), the Grayscale ETF application and exchange listing expectations, general AI sector hype, and its value as a long-term bet on decentralized AI. The $2.6 billion market cap is trading on these narratives, as the proven $3-15 million in annual revenue cannot support the valuation under any conventional framework.

Lecturas Relacionadas

Sequoia Dialogue with Jensen Huang: Computing Model Undergoes a 60-Year Transformation; You Won't Be Replaced by AI, But You Will Be Dimensionality-Reduced by 'Those Who Master AI'

NVIDIA founder and CEO Jensen Huang, in a conversation with Sequoia Capital's Konstantine Buhler, argues that we are witnessing the most significant computing shift in 60 years—from retrieval-based to generative computing. Instead of just storing and retrieving data, future systems will generate highly personalized content (text, images, video) on demand, powered by massive "AI factories." Huang envisions a global "intelligence network" that will envelop the planet, following the historical patterns of energy and communication grids. He outlines a five-layer investment framework: 1) Energy, 2) Chips/Computers, 3) Infrastructure (data centers), 4) AI Models, and 5) Applications. He predicts this ecosystem will reach a scale of $20 trillion annually. Crucially, Huang pushes back against fears of AI-driven job loss. He distinguishes between specific "tasks" (e.g., typing, analyzing images) and overall "jobs" (e.g., CEO, radiologist). While AI automates tasks, it increases efficiency and demand for the higher-value problem-solving aspects of professions, thus creating more jobs and "up-leveling" careers. The real risk, he asserts, is not being replaced by AI, but being outperformed by someone who effectively leverages it. He urges everyone to embrace AI as a tool for augmented capability and innovation.

marsbitHace 5 min(s)

Sequoia Dialogue with Jensen Huang: Computing Model Undergoes a 60-Year Transformation; You Won't Be Replaced by AI, But You Will Be Dimensionality-Reduced by 'Those Who Master AI'

marsbitHace 5 min(s)

"I Don't Need a Better Model Anymore": A Panorama of AI Users Under a Reddit Hot Post

Titled "I Don't Need a Better Model Anymore": AI User Reactions on Reddit Anthropic recently released Claude Fable 5, its first publicly available 'Mythos'-tier model, achieving 80.3% on the SWE-Bench Pro benchmark and significantly outperforming its predecessor and competitors. However, a viral Reddit post titled "Claude Fable made me realize I don't need better models anymore" highlighted a growing user sentiment of "good enough." Top comments expressed "model fatigue," with users stating that earlier models like Opus 4.5/4.8 already sufficed for their workflows. High cost was a key concern, as Fable 5's API is nearly twice the price of Opus 4.8, with users questioning the return on investment and suggesting the field has hit a plateau. The most frequent complaint targeted Fable 5's stringent safety filters. Designed to intercept high-risk requests (e.g., cybersecurity), the system was perceived as overly conservative. Users reported frequent rejections for routine security-related tasks, leading to automatic fallbacks to the older Opus model. Paying users were particularly frustrated, feeling they paid a premium for a less usable product. Dissenting voices came from users with heavy, complex tasks. For workloads like high-energy physics simulations with thousands of code lines, Fable 5's improved long-context understanding and error detection represented a significant, worthwhile leap—described as moving from a "college player to an NBA starter." The debate underscores a divergence between benchmark performance and practical utility. For most users, current models meet their needs, making further advances relevant only for extreme use-cases. The discussion also raised concerns about a potential "Public AI Freeze," where the most powerful models (like the restricted Mythos 5) remain exclusive to enterprises and governments, while public offerings stagnate. The launch presents two report cards: one of technical excellence and another of user skepticism. Fable 5's ultimate reception may depend on Anthropic's ability to refine its safety filters and justify its cost for specialized, high-demand users.

marsbitHace 12 min(s)

"I Don't Need a Better Model Anymore": A Panorama of AI Users Under a Reddit Hot Post

marsbitHace 12 min(s)

When AI Traffic Surpasses Humans, How Do You Prove You're Human?

With AI-generated web traffic surpassing human activity, websites face a crisis as AI agents bypass ads, avoid clicks, and scrape data without generating revenue. This disrupts the ad-based internet economy, diverting traffic and reducing site visits. In response, sites are blocking AI crawlers and deploying traps like Cloudflare's "honeypot" pages. Traditional CAPTCHAs are now ineffective against advanced AI. The focus has shifted to behavioral biometrics—analyzing unique human patterns such as cursor movement, typing rhythm, and keystroke dynamics. Companies like IBM and BioCatch use this data to distinguish humans from bots, even detecting fraud through behavioral inconsistencies. Two competing approaches aim to verify human identity centrally. Sam Altman’s World (formerly Worldcoin) uses iris scanning to create unique credentials, though it faces privacy concerns and regulatory bans. Alternatively, cryptographic zero-knowledge proofs offer anonymous verification without revealing personal data, championed by Vitalik Buterin to avoid centralized surveillance. However, both systems have flaws. Centralized solutions risk biometric data misuse, while decentralized models may be exploited through identity rental markets in economically unequal regions. Despite challenges, the author favors cryptographic methods for preserving privacy over pervasive behavioral monitoring that permanently captures and controls personal biometric data.

marsbitHace 21 min(s)

When AI Traffic Surpasses Humans, How Do You Prove You're Human?

marsbitHace 21 min(s)

2026 Landscape of Decentralized AI: Why is Blockchain the Inevitable "Antidote" for AI?

**The 2026 Landscape of Decentralized AI: Why Blockchain is the "Cure" AI Cannot Ignore** Decentralized AI addresses fundamental bottlenecks of centralized AI: scarce and expensive computational resources, excessive control concentration, unverifiable model outputs, and increasing difficulty in acquiring training data due to privacy and regulation. Blockchain offers a path to make intelligence open, verifiable, and economically accessible. The technical stack comprises three layers: 1. **Applications & Services**: The main crypto use cases are "Agentic Finance" (converting natural language into on-chain actions) and "Agentic Payments" for machine-to-machine commerce. Projects like Giza, Infinity Labs, Coinvest AI, and x402 (handling 173M+ transactions) are key players. 2. **Middleware**: This coordination layer enables agents to discover, identify, and transact. Notable projects include Gokite AI (specialized L1), Virtuals (an OS for the agent economy), and especially Bittensor—a network of specialized subnets forming competitive AI micro-economies. 3. **Infrastructure**: The capital-intensive layer providing raw resources. It includes decentralized compute (Akash, Render, Aethir), verifiable inference (Venice AI, OpenGradient), distributed training (Prime Intellect, Templar AI), decentralized storage (Filecoin, Walrus), and privacy/verification layers (Nillion, Arcium, Phala Network) using technologies like ZKPs, MPC, and TEEs. The outlook for 2026-2027 indicates AI demand outpacing infrastructure, with AI agents as a primary growth engine. Computation is becoming an asset class, with on-chain markets as its financial layer. Tokenomics is emerging as a structural advantage for coordinating capital, compute, and data in decentralized AI networks. While still early—with adoption uneven and revenue often trailing token incentives—projects like Bittensor, NEAR, and Virtuals demonstrate a shift from speculative narrative to a new model for coordinating intelligence.

marsbitHace 24 min(s)

2026 Landscape of Decentralized AI: Why is Blockchain the Inevitable "Antidote" for AI?

marsbitHace 24 min(s)

a16z Crypto Partner: Cash Flow is the Moat

Cash Flow as the Moat: A Playbook for Crypto Founders Historically, the most enduring businesses have been built by positioning themselves within the "flow of funds"—facilitating the creation and transfer of value in a network and extracting a portion of it. Cryptocurrency is the first modern technology natively built for this purpose. For startups, failing to architect products and businesses to leverage these principles means missing a major opportunity. Blockchains are inherently network businesses. Each transaction settles on a shared ledger, and every new participant strengthens the underlying network for all. Well-designed network tokens amplify this by aligning users, developers, and validators around growing the network, with value flowing back to contributors in a transparent feedback loop. This model is not new; companies from railroads and Standard Oil to Google, Meta, and AWS have thrived by inserting themselves into critical flows of value (goods, attention, compute). Financial markets make it even clearer: firms like Visa and major market makers generate immense revenue not by predicting markets but by being in the path of transactions. The combination of fund flow and network effects creates one of the most durable business structures. The high margins in traditional finance (payments, custody, lending, FX) represent prime targets. Crypto founders have the opportunity to build the next version—programmable, instant, global, and natively in the flow of funds. The frontier extends beyond finance to areas like computing/GPUs, AI training data, energy, robotics, and space—markets without entrenched intermediaries, ripe for building new, efficient value rails on programmable infrastructure. Founders should ask: Are you in the flow of funds today? Does your revenue scale 10x with the value of activity on your platform? Where in your target market are profit margins highest relative to value created? The opportunity is clear: embed your startup into the new flows of value and let the network effects accumulate.

marsbitHace 26 min(s)

a16z Crypto Partner: Cash Flow is the Moat

marsbitHace 26 min(s)

Trading

Spot
Futuros
活动图片