Cortex (CTXC) cryptocurrency is a decentralized Artificial Intelligence blockchain system, the network’s main objective is to make AI technology available for smart contracts. Using Cortex Smart AI Contracts, users can develop AI DApps (Artificial Intelligence Decentralized Applications) and also contribute towards other contracts^[1][5]. 

Cortex adds AI algorithms to support for smart contracts so that anyone can add AI to their smart contracts^[6].

Overview

Furthermore, Cortex has a mechanism for collective collaboration that permits everyone to submit and optimize models in Cortex. Those users that will be contributing to the models will be rewarded for their work. Cortex has created and developed something that is known as Artificial General Intelligence, also known as AGI^[2][3]. 

Cortex also has an Intelligent Inference Framework through which blockchain researchers globally upload researched data models in the storage layer of the Cortex network. Then other users who want to use these AI models can inference the stored data at a price^[7].

By inferencing the AI models, it makes the work of the nodes quite easy when a smart contract transaction is initiated since the nodes only need to inference the AI instructions from the stored models and coming to a consensus on the results. This way smart contracts can be executed at full nodes on a normal computer CPU^[1]. 

Cortex main mission is to provide state-of-the-art machine-learning models on the blockchain in which users can infer using smart contracts on the Cortex Blockchain. Moreover, Cortex wants to implement a machine-learning platform that would allow users to post tasks on the platform and AI DApps.

Cortex will provide a submission interface for training off-chain. At the same time, it will include the introduction-interpreting virtual machine for models. This will create a bridge between computing power and algorithm providers for trading and for collaborations^[2]. 

Cortex Features

Cortex’s core features include:

Smart AI Contracts: Cortex allows users to write machine learning programs on the blockchain, then submit interactions that are dependent upon other contracts.

Cortex Virtual Machine: Cortex has its own virtual machine called the Cortex Virtual Machine, or CVM. The CVM instruction set is fully Ethereum Virtual Machine (EVM) compatible and supports inference instructions.

Cortex Inference Consensus: When a user initiates a transaction to a contract, the full node needs to execute the code of the smart contract. Cortex smart contracts differ from ordinary smart contracts because Cortex’s “intelligence contracts” can be equipped with inference instructions, and then all nodes need to agree on the result of this inference. This creates consensus.

Model Submission Framework: Cortex also provides a submission interface for training off-chain, including submissions for virtual machines that interpret instructions for models. This creates a bridge between computing power providers and algorithm providers, allowing for trading and collaboration between participants in the Cortex ecosystem.

Cortex Intelligence Inference Framework: Machine learning researchers around the world can upload well-trained corresponding data models to Cortex’s storage layer. Then, users who need these AI models can make inferences using models, then pay the providers^[4]. 

The Team

Cortex is led by Ziqi Chen (CEO), who co-founded Waterhole.io with Cortex’s Chief Blockchain Engineer, Yang Yang. Other listed members of the team include Weiyang Wang (CTO), Xiao Yan (Chief Deep Learning Engineer), and Amy Chen (Operating Officer).

Cortex has received investments from Bitmain, FBG Capital, ZhenFund, DFund, Connect Capital, Blockwater Capital, Nirvana Capital, and other companies (most investments occurred during a private token sale in February 2018). They’ve also partnered with Tsinghua University, the University of California Berkeley, and Shanghai Jiao Tong University^[4].