What Is Bittensor (TAO) and How It Changes AI Development

What Is Bittensor (TAO) and What Goals Does It Have?

Bittensor is a blockchain platform focused on the open development of AI systems, providing easy access to resources and model training.

But why is this needed? 

For AI applications like ChatGPT or Midjourney to analyze information and generate responses, they are trained on large amounts of data, including images, text, videos, and other types. Models learn to recognize patterns and perform different tasks. This work requires large-scale computational resources, which in turn is associated with high costs. 

Mainly, the vast amount of data and resources is accessible to a limited number of large corporations. Meanwhile, for startups and smaller corporations, creating their AI models is a significant challenge.

As one of the projects focused on the intersection of AI and blockchain, Bittensor aims to solve this problem. Bittensor’s mission is to ensure that the benefits and ownership of machine intelligence are accessible to everyone. Operating as a shared computer, the platform allows developers to create AI applications on the blockchain and share resources, enabling AI models to learn from what other models have already been trained on.Bittensor is not owned by a single entity and is available for everyone to run, govern, contribute to, and earn money from their created tech products. The platform’s native token, TAO, bridges users with the platform, powering the network’s operations.

How Bittensor Works? 

The Bittensor blockchain supports different sub-networks (subnets), which run independently but are connected to the main blockchain (subtensor) and other subnets. Each subnet is associated with a certain purpose and machine learning model. 

For example, one of the key subnets is the marketplace for AI outputs. Based on their training and capabilities, AI models respond to requests, which are then checked by additional machine learning models. This mechanism allows AI models to adapt and learn from each other. There are subnets designed for different types of applications, like financial market prediction, storage space, and more. 

In addition to existing subnets, developers can build new ones to compete with others and get rewards in TAO tokens. Subnets experiment with different approaches and designs to improve their technologies. 

The Bittensor blockchain records all of the key activities of all the subnets. Relying on its Yuma consensus mechanism, the blockchain determines reward distribution for each subnet’s participants based on their contribution to the network. 

Bittensor architecture. Source: bittensor.org

TAO Token Price, Use Cases, and Tokenomics 

TAO is used to pay network participants for their computational resources, governance, transaction fees, and other purposes within the ecosystem. The token was launched in January 2021 along with the Bittensor network. TAO tokens were distributed among the network participants based on their contribution without an ICO and pre-mined tokens. 

Sharing Bitcoin’s principles of tokenomics, TAO also has a total supply of 21 million tokens. New tokens are produced through mining and block miners and validators get rewards for their work. After every 10,5 million blocks, the block rewards are reduced by half. TAO’s first halving is supposed to happen in August 2025. 

At the time of writing, the TAO price stands at around $555. TAO is the 26th largest cryptocurrency with a market cap of over $4 billion, and its daily trading volume is over $172 million with 7,380,936 TAO tokens in circulation. TAO reached its all-time high of $767 on April 11, 2024, and an all-time low of $30.4 on May 14, 2023.

Who Are the Founders of Bittensor

The co-founders of Bittensor are Jacob Steeves and Ala Shaabana are both Machine Learning researchers and software engineers. 

Based on data from Crunchbase, Bittensor was founded in 2019 in Canada. Behind the platform’s development is the Opentensor Foundation, which focuses on building open and accessible AI technologies.