Cortex is a public blockchain infrastructure that supports the execution of AI algorithms and AI-powered dApps. Coherently, machine-learning models would be ported onto smart-contracts where they can be called by users (i.e., "infering").
Cortex blockchain’s key technological milestone is its Cortex Virtual Machine (CVM). The CVM is ported from the Ethereum Virtual Machine (EVM), making it backward-compatible with the EVM. In addition, it features added layers for AI inference (infer instructions and deterministic inference engine). Cortex further uses a Proof-of-Work (PoW) consensus mechanism.
The CTXC token is used as a fee for transactions and smart-contract execution. When a transaction is invoked, Endorphins fees (equivalent to Ethereum’s gas) are allocated to miners and used to pay the AI / ML model providers.
1. What is Cortex (CTXC)?
Cortex is a public blockchain infrastructure that supports the execution of AI algorithms and AI-powered dApps. Cortex provides a platform for AI developers to upload their models onto the blockchain and incorporate them into smart contracts.
Cortex uses PoW network consensus to verify every step of the machine-learning inference.
The key technology of the Cortex blockchain is the Cortex Virtual Machine (CVM); the CVM is ported from the Ethereum Virtual Machine (EVM) and is backward-compatible with the EVM with added layers for AI inference (infer instructions and deterministic inference engine). Infer instructions allows models to be called in smart contracts through instruction sets. The deterministic inference engine, called Synapse, guarantees exactly the same result of an AI model in different computing environments and thereby making network consensus on AI program executions possible.
The CTXC token is used as a fee for transactions and smart-contract executions. When a transaction is invoked, Endorphins fees (similar to Ethereum’s Gas fee) are not only allocated to miners who help package the block on the full node but also used to pay the AI model providers.
A single execution of a contract will cause a fee of Endorphins Limit * max Endorphins Price. Generally speaking, the cost of the Endorphin is proportional to the size of the model. As a result, smart contract developers pay and sustain AI developers, while AI developers are incentivized to provide more quality AI models to smart contract developers, creating an ecosystem that leads to the evolution of better and better AI models. The Cortex blockchain can decentralize AI by creating a tokenomic model that incentivizes AI developers to share their models.
Anywhere a traditional application can be enhanced by AI, a decentralized application can be enhanced by AI. For instance, in gaming, on-chain AI can serve as impartial referees (such as in the game Digital Clash, recently released on Cortex) or enhance game characters (such as more life-like Crypto Kitties).
2. Economics and supply
Cortex raised funds through a Private Sale, which was conducted in February 2018 for 60MM CTXC at a rate of ~$0.53 per token and raised a total of 40,000 ETH (~$32.00 MM), selling 20.01% of the total token supply.
CTXC token distribution (%)
The following chart represents the number and breakdown of all CTXC that are to be released into circulation on a monthly basis.
Mining reward halves every 4 years, similar to the halving schedule of Bitcoin. Total mining reward is 150,000,000 CTXC, yielding a total cap of 299,792,458 CTXC. The mining process of CTXC will take place over 20 years.
CTXC token allocation schedule
3. Cortex's activity
Cortex currently focuses on community building across several geographies including Mainland China, Taiwan, South Korea, Thailand, Singapore, Russia, Canada, USA, and Latin America.
Cortex's current community development strategies include:
Publishing articles that explain AI and blockchain for the general public.
Attending and hosting workshops, meetups, conferences, and hackathons to engage with developers and investors.
Hosting bi-weekly technical/community updates.
Cortex's future community development strategies include:
Hosting online and offline development workshops.
Developer bounty programs.
Building open model libraries and datasets.
Collaborating with academia for research partnerships and publications.