Cortex is an open source public blockchain platform that aims to bridge the gap between artificial intelligence and blockchain technology. Its primary mission is to create a decentralized artificial intelligence system that allows machine learning models to be uploaded and executed on a distributed network. By doing so, Cortex enables the integration of AI models into smart contracts, which allows developers to build more advanced and intelligent decentralized applications, often referred to as AI DApps. A core component of the platform is the Cortex Virtual Machine. While it is compatible with the Ethereum Virtual Machine, it is specifically designed to handle the computational requirements of machine learning by utilizing the GPU rather than just the CPU. This allows the network to perform complex AI inferences directly on the blockchain in a deterministic way. To ensure that AI results remain consistent across different nodes and environments, the project uses a specialized inference engine called Synapse. The project also features a technology called MRT, or Material Representation Tool, which helps in quantizing and optimizing AI models for efficient execution. Additionally, Cortex has developed a Layer 2 solution known as ZkMatrix. This solution uses zkRollup technology to aggregate transactions, which helps increase the overall speed of the network and reduces the computational cost associated with executing contracts. The native token of the network, CTXC, serves as the primary utility token within this ecosystem. It is used to pay for transaction fees and the execution of smart contracts. These fees are often referred to as Endorphins, functioning similarly to gas on other networks. A unique aspect of the Cortex ecosystem is its incentive model for AI researchers and developers. When an AI model is uploaded to the blockchain and subsequently used by a DApp or a smart contract, the original developer of that model is compensated. This creates an open marketplace where contributors are encouraged to share and optimize high quality machine learning models. The consensus mechanism used by the network is based on a Proof of Work algorithm that is optimized for AI tasks, ensuring security while accommodating the platform's high computational needs. The project was founded by individuals with backgrounds in large scale systems and research, including Ziqi Chen and Jia Tian. Overall, Cortex positions itself as a decentralized world computer capable of hosting and running artificial intelligence without relying on centralized off chain solutions.
Learn more
