Research
Institutional-grade research on the Bittensor ecosystem - monthly updates, subnet deep dives, and investment theses from the TAO Institute team.
Monthly Update on TAO Ecosystem
TAO Ecosystem Update - February 2026
The TAO ecosystem saw significant momentum in February with three new subnet registrations, a 12% increase in total staked value, and the rollout of dynamic TAO (dTAO). We break down the key metrics, governance proposals, and emerging trends shaping the network this month.
Read Report →TAO Ecosystem Update - January 2026
January kicked off with record-breaking miner participation and a surge in subnet revenue. Our monthly recap covers the state of emissions, validator consolidation trends, and the growing institutional interest in TAO staking products.
Read Report →TAO Ecosystem Update - December 2025
December's ecosystem review examines end-of-year network health, the impact of halving expectations on TAO price, and a retrospective on the top-performing subnets of 2025. Plus: our predictions for 2026.
Read Report →TAO Institute Subnets
Subnet 1: Text Prompting (Apex) - Deep Dive
The flagship text generation subnet and the backbone of Bittensor's inference network. We analyse Apex's architecture, competitive moat, miner economics, and why it consistently commands the highest emissions allocation.
Read Report →Subnet 8: Taoshi - Trading Intelligence Analysis
Taoshi has emerged as the leading DeFi-focused subnet, leveraging decentralized AI for quantitative trading strategies. This report covers its proprietary signal architecture, revenue model, and competitive positioning against centralised quant funds.
Read Report →Subnet 9: Pretrain - Foundation Model Training at Scale
Pretrain enables distributed pretraining of large foundation models across the Bittensor network. We evaluate the subnet's technical architecture, training efficiency metrics, and the long-term viability of decentralised model training.
Read Report →Subnet 13: Dataverse - The Decentralised Data Layer
Dataverse is building the data marketplace infrastructure for AI training pipelines. This analysis covers its indexing mechanisms, data quality guarantees, and strategic importance as AI models demand ever-larger datasets.
Read Report →Subnet 68: Nova - AI-Powered Trading Engine
Nova is the fastest-rising subnet by momentum, combining multi-model ensemble inference for market analysis. Despite a lower Khala Score, its rapid growth trajectory and novel architecture make it one to watch in 2026.
Read Report →Subnet 28: ZK Proofs - Privacy-Preserving Compute
The ZK Proofs subnet is carving a unique niche at the intersection of zero-knowledge cryptography and decentralised AI. We assess the technical complexity, market opportunity, and team execution in this emerging vertical.
Read Report →VC Thesis on Subnets
Investment Case for Decentralised AI Training
Centralised AI training is a $40B+ market dominated by hyperscalers. This thesis examines how Bittensor's subnet model could capture meaningful market share by offering cost-efficient, censorship-resistant alternatives - and what that means for TAO's long-term value accrual.
Read Report →Why Bittensor Subnets Are the New L2s
Ethereum L2s created a multi-billion dollar ecosystem by specialising execution. We argue Bittensor subnets represent an analogous opportunity in the AI stack - specialised intelligence layers with native economic incentives and composable value flows.
Read Report →TAO as a Portfolio Allocation: Risk-Adjusted Returns Analysis
For institutional allocators evaluating TAO exposure, we present a quantitative framework covering Sharpe ratios, correlation analysis against BTC/ETH, liquidity profiles, and optimal portfolio weighting scenarios based on historical and projected network fundamentals.
Read Report →The Subnet Flywheel: Network Effects in Decentralised Intelligence
Network effects are the most defensible moat in technology. This report maps the feedback loops between miners, validators, subnet owners, and stakers - and models how these dynamics compound TAO's competitive position over time.
Read Report →Bittensor's Moat: Why Competitors Can't Just Fork TAO
Open-source protocols face constant fork risk. We analyse the structural, economic, and social moats that protect Bittensor - from validator lock-in and subnet switching costs to the compounding advantages of the network's intellectual capital.
Read Report →From Subnets to Revenue: Modelling TAO's Path to Cash Flow
Crypto's perennial challenge is bridging from speculative value to fundamental cash flows. We model three scenarios for how subnet revenue, staking yields, and ecosystem fees could transform TAO from a narrative asset into a yield-generating protocol.
Read Report →