The decentralized AI project x.TAI based on Ordinals opens Phase 1 to solve a fundamental problem in the field of AI development: database shortage and uneven value distribution in the data labeling process. x.TAI's solution to this is to enable participants to provide a training dataset for a decentralized large language model (DeLLM) by annotating the data, and use the inscription $xtai on Brc-20 as a reward. Key highlights of x.TAI include: 1. It challenges the hegemony of centralization in the traditional data labeling process, such as CAPTCHA verification tasks, where centralized institutions obtain datasets without fairly compensating contributors. x.TAI plans to reward individuals who contribute to their data annotation process with $xtai. 2. DeLLM: x.TAI will use datasets to train decentralized large language models, so that all participants can share the ownership of the model, and adopt a decentralized approach to overcome problems such as data security vulnerabilities, resource monopoly, and misinformation. 3. AI deinfra: xTAI will lay the foundation for the decentralized development of AI and guide a more decentralized, fairer, and safer way of AI development around the world. Related Links: xtai.io