Educational material. Not financial advice.

Data as of June 11, 2026.

What is this about?

Bittensor is an open protocol designed to build a decentralized network for machine learning on the blockchain. The core idea revolves around an AI marketplace where miners provide computational resources and machine learning models, validators assess their quality, and participants earn rewards in TAO tokens proportional to the utility of their contributions. The project was launched in 2021, founded by Jacob Steves and Ala Shaaban. Since 2023, Bittensor has been actively developing a modular architecture of 'subnets', where each subnet specializes in a specific AI task.

Basic figures

Price: $210.33 Market cap: $2.26 billion FDV (fully diluted valuation): $4.41 billion 24h trading volume: $132.19 million Volume to market cap ratio: ~5.8% CoinMarketCap ranking: #34–37 (varies by source) ATH: $756.83 (March 7, 2024) Deviation from ATH: -72.2% ATL: $0.1972 (March 3, 2023) Circulating supply: 10.76 million TAO (51.3% of max) Maximum supply: 21,000,000 TAO Blockchain: proprietary (Substrate / Polkadot SDK) Explorer: taostats.io

How the product works

Network architecture

Bittensor is designed as an 'intelligence market' comprising three types of participants. Miners are nodes providing AI services: running models, responding to requests, generating predictions. Validators are nodes assessing the quality of miners' responses and forming a weight matrix (Yuma Consensus). TAO stakers are token holders delegating stake to validators.

Subnet architecture

Since 2023, the network has transitioned to a modular structure. Each subnet is a separate AI application with its own validation rules, its own miners, and its own utility assessment mechanism. As of the end of 2025, the network had 128 active subnets covering tasks from text generation to financial forecasting and bioinformatics.

Dynamic TAO and subnet tokens

In April 2026, the Dynamic TAO mechanism was introduced: each subnet now has its own token, and liquidity is integrated directly into the protocol. This expands the network's economy and allows the market to evaluate individual subnets independently.

Reward mechanism

TAO is emitted through the blockchain every 12 seconds. After the halving at the end of 2025 — 0.5 TAO per block (3,600 TAO per day). Distribution among subnets is determined by validator voting. In May 2026, the mechanism was updated: TAO emission is primarily concentrated in the most productive subnets, creating competition for quality.

Team

Jacob Robert Steeves — co-founder

Education and experience:

  • Bachelor's degree in Mathematics and Computer Science, Simon Fraser University (2011–2015)

  • Machine Learning Researcher, Knowm Inc. (2015–2016)

  • Software Engineer, Google (December 2016 — April 2018)

  • Co-founder of Bittensor / Opentensor Foundation since 2019

  • Based in Peru

Public activity:

  • Active on X (Twitter) as a technical speaker for the project

  • Regularly comments on protocol updates and publicly responds to criticism

Controversial situations:

  • April 2026 — public conflict with Samuel Dair (Covenant AI). Covenant AI — one of the largest subnet operators — announced its exit from the network, accusing Steeves of centralized control: according to Covenant AI, Steeves initiated 38 of 41 upgrades to the network during 2023–2026. Steeves himself called the accusations 'a deep betrayal' and publicly denied them. The event triggered a 25% price drop in a day and erased about $650 million in capitalization.

Ala Shaaban — co-founder

Data on current roles in public sources are limited. Mentioned as a co-founder in project profiles.

GitHub activity (Opentensor team)

Main repositories: opentensor/bittensor (SDK), opentensor/subtensor (blockchain layer), opentensor/bittensor-subnet-template. According to open sources, the repositories were updated in March–April 2026. Exact data on the number of commits and active contributors for the last 90 days is not publicly available as of the analysis date.

Regulatory status

The project has no publicly known solutions from the SEC or other regulators. Jurisdiction — not officially disclosed (Jacob Steeves is based in Peru). However, several analysts point out that the structure of delegation and TAO rewards could potentially fall under the criteria of the Howey test.

Funding

Bittensor fundamentally differs from most large crypto projects: traditional venture rounds and ICOs were not conducted. Tokens are not issued all at once — they are mined through participation in the network (analogous to Bitcoin mining). There are no classic vesting schedules for the team or investors.

Despite this, several institutional players have formed positions in the open market:

  • DAO5 (Tekina Salimi's fund, a former Polychain member) — position ~$50 million according to open sources

  • Digital Currency Group (DCG) — ~$100 million in TAO

  • Nvidia — according to MEXC News, formed a position of about $420 million, of which ~77% is staked

  • Polychain Capital — added ~$200 million in Q1 2026

Note: all figures regarding institutional players' positions are based on third-party sources (MEXC News, Blockonomi). No official confirmations from the companies are publicly available.

General Tensor (infrastructure provider for Bittensor) closed a pre-seed + seed round of $5 million — one of the few confirmed rounds in the ecosystem.

Supply distribution

Bittensor has no allocations in the classical sense (team / investors / ecosystem). All tokens are mined by participants:

  • ~100% — mined by network participants (miners, validators) through a mechanism similar to Bitcoin

  • Maximum supply: 21,000,000 TAO

  • Current circulation: ~10.76 million TAO (51.3%)

  • Emission rate after the first halving (end of 2025): ~3,600 TAO/day

  • Second halving: automatically upon reaching 15,750,000 TAO issued (expected around 2028–2030)

  • ~68–77% of circulating supply staked at the time of analysis (source: Blockonomi, MEXC News)

Project metrics

Declared by the team / published through official channels

  • 128 active subnets as of the end of 2025 — early 2026

  • Roadmap Robin: expansion to 256 subnets

  • Conviction v2 (BIT-0011): new governance mechanism with a time-lock on stake

Independently verified

  • $43 million — network revenue from real AI usage in Q1 2026 (source: Blockonomi, MEXC News — referencing on-chain data)

  • 34% growth — number of active validators in a month (March 2026, source: MEXC News)

  • $132–162 million — daily trading volume of TAO (CryptoRank, CoinMarketCap — independent aggregators)

  • $2.26B — market capitalization (CryptoRank)

  • 70%+ — share of staked supply (several independent sources indicate a range of 68–77%)

Important note

Data on DAU of subnets, the actual number of unique users, and independent verification of $43M revenue from specific sub-protocols in public access is limited. DeFiLlama aggregators classify Bittensor in a separate category (not a DeFi protocol in the classic sense), TVL in the traditional understanding is not applicable.

Market context

This section describes the market environment of the token. It does not contain trading signals, forecasts, or entry assessments.

Price dynamics

  • 24 hours: +2.22%

  • Q1 2026: +21.57%

  • From ATH ($756.83, March 7, 2024): -72.2%

  • From ATL ($0.1972, March 3, 2023): +106,000%+

Significant events chronologically

  • December 2025 — first halving of TAO: emission cut in half

  • April 28, 2026 — submission of ETF applications from Grayscale and Bitwise, prompting a price surge

  • April 2026 — conflict between Covenant AI / Jacob Steeves: -25% in a day, ~$650M in capitalization wiped out

  • SEC decision on ETF: expected no earlier than August 2026

Listings and liquidity

TAO is traded on the largest centralized exchanges: Binance, Coinbase, OKX, Bybit, Kraken. On decentralized platforms — through native subnet liquidity and DEX adapters. The volume to market cap ratio (~5.8%) is moderate. Futures on TAO are available on Binance, Bybit, OKX.

Competitive analysis

Bittensor positions itself as a decentralized AI market. The niche includes both infrastructure AI protocols and projects with a more narrow focus.

Artificial Superintelligence Alliance (FET/ASI) — cap ~$2–3 billion

  • Model: unified protocol of agents (Fetch.ai + SingularityNET + Ocean Protocol). Focused on AI agents and data marketplace.

  • Users / activity: DAU data are not systematically disclosed publicly; an active ecosystem of agents exists across several chains.

  • Stronger than TAO: broader coverage of use cases (data + agents + compute); powerful marketing and recognition of SingularityNET (Ben Goertzel).

  • Weaker than TAO: absence of a unified subnet mechanism and built-in quality economy; coordination of three merged organizations creates operational risks.

NEAR Protocol — cap $3.6–6.7 billion (varies)

  • Model: L1 blockchain, actively positioning itself as an AI blockchain through the narrative of 'Chain Abstraction' and on-chain AI integration.

  • Users / DAU: hundreds of thousands of transactions per day according to on-chain data.

  • Stronger than TAO: functioning L1 with a broad dApp ecosystem, high liquidity, established brand; does not depend on the AI narrative as the sole source of value.

  • Weaker than TAO: AI is a marketing positioning rather than an embedded mechanism for generating value; lacks a subnet architecture with a real market for AI services.

Render Network (RENDER) — cap ~$1–2 billion

  • Model: decentralized market for GPU computing. Utilizes underutilized GPUs for rendering and AI inference.

  • Users / activity: number of active nodes and volume of completed tasks according to the internal dashboard.

  • Stronger than TAO: specific, clear use case (GPU rental); less dependent on the narrative of 'AI in crypto'.

  • Weaker than TAO: lacks its own quality assessment of AI models; covers infrastructure but not the 'market of intelligence' as a whole.

Gensyn — non-public (pre-token)

  • Model: decentralized verification of ML training with cryptographic proof. Positioned as an infrastructure layer for AI training.

  • Users / activity: data in the testnet phase; token not issued.

  • Stronger than TAO: more rigorous cryptographic verification of computations; attracting institutional ML researchers.

  • Weaker than TAO: no functioning token market; ecosystem at an early stage.

Ritual — non-public / early stage

  • Model: on-chain AI inference layer, integration of AI models into smart contracts.

  • Stronger than TAO: focused on composable AI within DeFi protocols — a new niche not occupied by TAO.

  • Weaker than TAO: no token and public metrics; extremely early stage.

Substantive difference of Bittensor

The key structural difference of TAO from most competitors is the built-in quality assessment mechanism, not just the incentivization of computation supply. Validators assign weights to miners based on comparative quality of responses. Subnet architecture creates competing specialized markets. This forms an economic structure that is harder to replicate than a simple GPU marketplace or AI labeling on an existing L1.

Partnerships and ecosystem

Confirmed partnerships:

  • Yuma / DCG (2024–2026) — asset management for subnet tokens; institutional access to TAO and subnet derivatives

  • BitGo (2025–2026) — institutional custodian making TAO and subnet tokens accessible to institutional clients via a staking platform

  • Cerebras Systems — collaboration with an AI chip manufacturer in the context of scaling computations; integration details in public sources are limited

  • GenomesDAO — a subnet for AI bioinformatics tasks (including protein-folding models)

  • LayerZero — exploring the possibility of cross-chain interaction for subnet communications

Institutional products:

  • Grayscale Bittensor Trust — filed Form S-1 (December 30, 2025) to convert to a spot ETF under the ticker GTAO on NYSE Arca

  • Bitwise — joint (with Grayscale) application for a spot TAO ETF; SEC decision expected around August 2026

  • Grayscale AI Fund — increased TAO share to 43% of the fund's weight

Grants and ecosystem programs:

  • Opentensor Foundation supports independent subnet developers through grant programs. Details are publicly limited.

Key project development dates

2019

  • Start of protocol development by Jacob Steeves and Ala Shaaban

2021

  • November — mainnet launch, first exchange rate fixed at ~$90

2022–2023

  • Gradual network expansion, first specialized subnets

  • March 2023 — ATL: $0.1972 (according to CryptoRank)

  • 2023 — average price for the year $97.28, year-end closing ~$267

2024

  • March — ATH: $756.83

  • Introduction of subnet tokenomics, growth to 64+ subnets

2025

  • Growth to 128 active subnets

  • December — first halving of TAO: emission from ~7,200 to ~3,600 TAO/day

  • Grayscale files Form S-1 for GTAO spot ETF (December 30)

2026

  • Q1 — Nvidia ($420M) and Polychain ($200M) forming positions; the network generates $43M in revenue from AI

  • April — launch of Dynamic TAO, subnet tokens with integrated liquidity

  • April — conflict between Covenant AI / Jacob Steeves: -25% in 24 hours

  • April–May — implementation of BIT-0011 Conviction Mechanism (time-lock staking for governance)

  • May — update of the emission mechanism: concentration of TAO in top subnets

In development (according to team statements):

  • Robin upgrade: expansion limit to 256 subnets

  • Conviction v2 in governance: devnet-ready, testing

  • SEC decision on spot ETF (expected ~August 2026)

Sentiment and community

Platform activity

  • Discord: 49,189 members (official server)

  • Telegram: 6,249 members (official channel)

  • Twitter/X: official account @bittensor_ (opentensor); active discussions in the crypto community

  • Reddit and forums: active discussion, including critical

Market sentiment

The structure of Q1 2026 reflected two competing narratives: aggressive institutional positioning (ETF applications, nine-figure purchases by Nvidia and Polychain) and a destructive governance crisis. Both factors acted simultaneously, creating increased volatility.

The spot ETF narrative retains institutional investor interest even amid internal disagreements. The number of open positions in TAO futures (Open Interest) on major exchanges increased in Q1 2026 alongside the price.

Signs of growth vs hype

  • In favor of organic growth: $43M in real revenue from AI usage in Q1 — this figure appears in references to on-chain data, not just in team releases.

  • In favor of narrative: a significant portion of attention to TAO correlates with the overall 'AI crypto' narrative of 2023–2026, not just with the fundamental indicators of a specific protocol.

  • A high percentage of staking (68–77%) reduces real float and may create the illusion of a liquid market with lower actual turnover.

Risks

🔴 High: centralization of governance

In April 2026, a serious situation arose concerning control over the protocol. Covenant AI founder Sam Dair publicly accused Jacob Steeves of unilateral control over Triumvirate — the key governance body of the network — noting that 38 of 41 upgrades were initiated personally by Steeves. Bittensor positions itself as a decentralized network; however, the current governance architecture, according to the departing party, does not align with this thesis. BIT-0011 Conviction Mechanism aims to partially address the issue, but the mechanism is in testing.

🔴 High: regulatory uncertainty (Howey test)

The structure under which participants stake TAO and receive rewards based on the efforts of other network participants may potentially qualify as an investment contract under the Howey test by regulators. The SEC has not yet released an official position on TAO. The decision on the spot ETF in August 2026 will serve as an indirect signal of regulatory attitude towards the token.

🔴 High: risk of ETF rejection

The spot ETF narrative became one of the key drivers in Q2 2026. Rejections of Grayscale and Bitwise applications or prolonged delays could create significant pressure, as part of the current pricing reflects expectations of approval.

🟠 Medium: quality and real usefulness of subnets

128 subnets is a quantitative indicator. The quality of AI services they provide and the real demand for these services outside the internal ecosystem is not publicly verified by independent analysts. Revenue of $43M in Q1 2026 is mentioned in several sources; however, detailed breakdowns by subnet and independent audits of the data are not available publicly.

🟠 Medium: concentration of large holders

Nvidia ($420M), Polychain ($200M), DCG (~$100M) — three holders with a total position of around $720M against an overall capitalization of $2.26B. This creates a risk of selling pressure, although a high level of staking (70%+) limits some of this risk.

🟠 Medium: competition with centralized AI providers

OpenAI, Anthropic, Google Gemini, xAI provide AI services with quality assurance, speed, and integration tools unattainable for a decentralized network in the short term. Bittensor's advantage lies in the absence of censorship, openness, and tokenized incentives, but this is not always a priority for the mass user.

🟡 Moderate: technical scaling risks

Transition to 256 subnets, implementation of Conviction v2, update of emission economy — all of these are major changes to the protocol. Mistakes in implementation or unforeseen interactions between mechanisms may lead to malfunctions. The Substrate architecture (Polkadot SDK) imposes its limitations on the speed of iterations.

🟡 Moderate: reputational risks of the team

The conflict with Covenant AI received widespread coverage. If similar situations recur or accusations of centralization find further confirmation, it could impact the trust of developers and subnet operators in the protocol.

🟢 Low: inflation risk of supply

Endless emission is not possible: hard cap of 21M tokens, progressively decreasing emissions through halvings, high staking levels — all this structurally limits inflationary pressure compared to most competitors.

Development scenarios

Description of the conditions under which each scenario may materialize. It is not an assessment of probability or a recommendation.

Bull scenario — conditions for realization:

  • SEC approves spot TAO ETF in August 2026, opening institutional demand through traditional investment products

  • Conviction v2 governance is perceived by the community as a compelling solution to decentralization issues — developer outflow is halted

  • Expansion to 256 subnets accompanied by growth in real revenue (independently verifiable data)

  • The AI narrative in the cryptocurrency sector persists or intensifies against the backdrop of technological events in 2026–2027

  • Confirming signals: growth in subnet revenue quarter-on-quarter; new major subnet operators; official SEC statements

Base scenario — inertial development:

  • ETF is approved, but without a frenzied influx — moderate institutional interest

  • Governance issues are partially addressed through BIT-0011, but criticism in the community persists

  • The number of subnets is growing, revenue is slowly increasing, competition with centralized AI prevents expansion beyond the crypto audience

  • The price of TAO remains within a range determined by current capitalization and overall market dynamics

  • Confirming signals: stable number of active subnets, moderate growth in on-chain activity, absence of new major scandals

Bear scenario — conditions for realization:

  • SEC rejects both ETF applications or postpones the decision to 2027+

  • Internal conflict over governance remains unresolved; major developers or validators continue to leave the network

  • Quality checks of subnets reveal that most do not produce genuinely demanded AI services — the narrative of 'intelligence market' loses trust

  • Regulatory decision on TAO as a security (SEC, other jurisdictions) creates legal limitations

  • Confirming signals: decrease in the number of active validators; decline in subnet revenue; exit of key institutional holders

Open questions for further observation

  1. Will Conviction v2 (BIT-0011) satisfy the community as a real solution to the centralization problem, or will accusations against Jacob Steeves continue to affect operator outflow?

  2. What is the real composition of the $43M revenue in Q1 2026 by subnets — are there several major generators or is this widely distributed data?

  3. Will the emission tokenomics withstand increased demand for TAO upon ETF approval, given that 70%+ of the supply is locked in staking?

  4. What will Nvidia and Polychain's response be to the unlocking of potential positions — will they hold or use it as liquidity?

  5. What will happen to the economy of subnets after the Robin upgrade (256 limit) — will the expansion dilute TAO emission across weak subnets, reducing incentives for quality operators?

  6. How Dynamic TAO and subnet tokens integrate with the spot ETF narrative — are only TAO considered in the ETF or potentially the entire range of subnet assets?

Sources

CryptoRank, CoinMarketCap, CoinGecko, Blockonomi, MEXC News, CryptoTimes, AInvest, IQ.wiki, TheBlock, Motley Fool, Tracxn, CoinStats AI, TaoStats, Yahoo Finance, Bitcoin.com News, Cryptopolitan, ChainCatcher, CryptoDaily

This material is a fundamental analysis of a cryptocurrency project for educational purposes. It is not financial, investment, or trading advice. It is not an offer to buy or sell assets. Cryptocurrencies carry a high risk of capital loss. Any decisions regarding asset actions are made by the reader independently and at their own risk.

#TAO