Now, if you’re part of an organization and want in on the action, there are two doors to choose from. First, the Open-Source archive, where you share data for everyone to use. Second, the private dataset pathway for those hush-hush files that help train AI but stay under wraps.