Crowdfunding and Crowdcomputing for Open-Source AI Models

A Reddit discussion has proposed using crowdfunding and crowdcomputing platforms to finance the training of large open-source language models like Qwen. The proposal suggests that both individual users and companies could contribute financially to reach milestones for training larger model sizes. Training state-of-the-art AI models requires massive financial resources and computational power, which are currently dominated by tech giants. Crowdfunding could democratize AI development by allowing the community to directly fund and own open-source alternatives. While the idea is appealing, training large models requires coordinated high-bandwidth clusters, making decentralized crowdcomputing technically challenging compared to centralized cloud providers. Additionally, managing trust, fund allocation, and licensing for crowdfunded models presents significant organizational hurdles.