Roblox downloaded 145 GB of models from our sites without asking (open3dlab.com)

In a controversial move, Roblox's crawler recently downloaded 145.5 GB of 3D models from creators on Open3DLab, SFMLab, and SmutBase without prior consent. Over two days, it scraped 712 projects, including adult content, in violation of the sites' Terms of Service and licensing agreements. This incident raises significant concerns regarding data ethics, as it appears Roblox may be collecting data for its generative AI systems, specifically its ongoing development of a "3D foundation model." The company has faced criticisms before for its treatment of content creators and its practices related to copyright and user safety. The technical implications are stark; the scraped models, many of which are licensed for non-commercial use, could inadvertently introduce adult content into Roblox's offerings aimed at children. This situation underlines the growing tension in the AI/ML community regarding copyright infringement and the ethical dimensions of scraping data without permission. With Roblox's previous legal troubles over content rights, this incident further highlights the urgent need for clearer regulations surrounding AI data sourcing and creator rights, and it reflects a broader trend of AI companies engaging in data collection practices that risk breaching ethical standards.