Idea

[Why] Issue:

  • Costs: Training AI models requires a large amount of data and computing power, mainly for large corporations and research institutes due to high costs. It is very difficult for small companies or startups to access. Additionally, the models are increasingly complex and costs will continue to rise.

  • The concentration of AI in the hands of a few organizations leads to risks of manipulation and abuse of AI for malicious purposes.

  • Collaborative development: Most AI models are kept private, limiting collaboration => Isolated models cannot learn from each other => Hindering the rapid development of the AI space.

[What] Idea: Bittensor identifies itself as a 'Decentralized Machine Learning' platform; they use the superior properties of blockchain to create a decentralized Machine Learning network where all parties can easily access, collaborate, share computing resources and data, thereby reducing costs and barriers when building complex machine learning models. The ultimate goal is to open new frontiers in AI advancement by synthesizing the knowledge contributions of each individual.

In simpler terms, Bittensor builds a vast AI network that includes all important actors and items in the AI market (including: data, storage, computing power,....)

[How] To achieve this, Bittensor has been built with a structure of one main blockchain and multiple interconnected subnets (similar to the structure of Avalanche) + a reward mechanism to encourage the participation of developers and organizations.

The structure and operational model of Bittensor:

The structure of Bittensor consists of three main components: Subnet, blockchain, and Bittensor API. Each subnet on Bittensor includes two main components: miner and validator. The way it works:

Overall, the operational model of Bittensor is relatively similar to $BTC Bitcoin, but users perform AI tasks instead of trading on the network.

Bittensor's reward mechanism creates a system that values talent, where the quality and utility of contributions are directly linked to economic rewards. This encourages participants to provide accurate, valuable data and models, as higher quality contributions will receive more tokens. This naturally discourages low-effort or malicious contributions, as they are less likely to be valued by the network.

Bittensor's differentiator:

  1. The project claims that the core technology improvement of Bittensor is the separation between the core operations of the chain (such as transferring money) and the validation systems used to create markets for its digital goods. This is a distinguishing feature compared to other blockchain systems like Bitcoin, Ethereum, and Filecoin.

=> Benefits:

  • Bittensor does not integrate validation systems into the blockchain, allowing these systems to be written in any programming language (Rust, Python, C++,...).

  • The necessary tools for validating subnet mechanisms are off-chain, allowing for heavy data processing and complex calculations.

=> Risks:

  • Lack of reliability: When the validation and validator systems operate off-chain, there is no reliable way to ensure that validators or miners are not cheating.

  • Lack of Process Recording: There is no record of the processes or calculations that have taken place to produce a specific outcome. This makes miners/validators 'black boxes' to each other, unable to verify or audit the processing steps.

Compared to Ethereum:

  • Bittensor is not a contract platform on-chain. It uses an off-chain activity scoring mechanism, with scores stored on the blockchain. This means that Bittensor's reward mechanism is less transparent and harder to trace compared to Ethereum.

  • The contract code of Ethereum is completely transparent and can be verified by users to ensure fairness and accuracy, while Bittensor's reward mechanism, on the contrary, is not clear. The processes determining rewards are not visible or fully audited.

  • Ethereum allows users to interact with the platform without needing to trust an intermediary, as everything is encoded in smart contracts that anyone can check. Bittensor, on the other hand, requires users to trust the fairness of the system due to the lack of transparency in how rewards are calculated and distributed.

Some operational results of Bittensor:

  • 49 Subnets are currently active since the launch started in October 2023 (Subnet slots are limited to 120)

Some notable Subnets: Myshell (raise $11M), Kaito AI (Web3 data research), NOUS, Taoshi, Corcel, Protein Folding.

=> Overall, Bittensor's subnets are starting to become quite diverse, but still lag far behind the top players in their product niche in the traditional market.

  • 31 Validators are staking more than 5.9M TAO tokens - accounting for nearly 80% of the current circulating supply. Among them, Polychain is the fifth-largest validator with over 510K $TAO tokens.

Team and Investors

  1. Team

Bittensor is a project developed by the Opentensor organization:

  • Jacob Robert Steeves: Co-Founder of Opentensor, many years of experience as a software engineer at large enterprises like Google, Knowm.

  • Ala Shaabana: Co-Founder of Opentensor, he has been an engineer at several companies such as VMware, Instacart...

  • James Woodman (COO): expertise in finance and business, has worked at many large enterprises such as Growth Equity at Mercato Partners, Investment Banking – FinTech M&A at Perella Weinberg Partners, and even GSR investment fund in a Business Development role.

  1. Investors

Invested by Digital Currency Group (one of the leading funds in the market, behind Grayscale), Polychain Capital, GSR, and FirstMask (a New York fund that early invested in Pinterest and invested in top hot projects like Discord, Airbnb, Riot Game, Shopify, Draftking,... - The founder has been recognized by Forbes for 5 consecutive years as one of the top 100 highest-return investors)

Additionally, Grayscale has launched a fund certificate for Bittensor, and Pantera has also invested (unverified source)

Tokenomics

TAO is a token designed similarly to Bitcoin, with a maximum total supply of 21,000,000 and gradually distributed in the form of block rewards for miners and validators. Each block receives a reward of 1 TAO.

Additionally, TAO also has a halving cycle that occurs every 4 years, with the block reward being cut in half. The first halving is expected to take place in November 2025.

Every 12 seconds, 1 TAO will be released, corresponding to a figure of 7200 TAO distributed to miners, validators, and subnet owners daily at a rate of 41%, 41%, 18%.

  • Total Supply 21,000,000 $TAO

  • Circulating Supply: 7,509,000 $TAO (Over 5.9M tokens account for nearly 80% of the current circulating supply are being staked - the annual inflation rate of TAO is at 28%)

  • CAP: 4B

  • FDV: 11.5B

  • Listed on almost all CEX exchanges, so liquidity is very strong, with an average daily volume > 50M$

Token Use Case:

Some important notes about Bittensor Eco:

  • Each Subnet will have its own token, this token is paired with TAO in the liquidity pool.

  • The higher the Token's market capitalization, the more emission (TAO reward) that Subnet receives.

=> This leads to miners having less incentive to sell when they receive subnet tokens because selling would reduce their TAO rewards for their subnet.

Onchain:

  • Binance is currently the largest holder of TAO with nearly 566k (accounting for about 35% of the circulating tokens aka tradeable). Mexc is in second place with 138k $TAO.

  • 107k TAO holders - quite a large number.

Upcoming update of Bittensor

Preparing to launch smart contracts and be compatible with Ethereum.

Community:

  • Twitter 77k followers: followed by top founders in the industry, posting quite frequently and interacting at a good level.

  • Discord 30k members - a pretty good number (half of Near, more than Polkadot)

  • Looking at Github, it can be seen that the dev team is still actively working.

=> Overall, the community is quite stable for an Infra project, although the project's product is mainly targeting a small number of users in the market (developers), not end-users yet.

Conclusion:

  • Bittensor's vision and goals are very ambitious, but their journey has just begun. It is necessary to closely observe each development step of the project.

  • Bittensor's strengths lie in the team, a strong group of investors + being highly regarded and supported by many in the industry.

  • $TAO is currently highly valued, and the token has a large inflation rate. However, the positive point is that the circulating supply is being tightened, which can easily drive prices up quickly, especially when AI becomes hotter than ever.

#AI #BTC