If I read it correctly, YaCy is like using a torrent except the information being shared between peers is search information. When you do a search you’re essentially asking everyone else in the swarm for access to their YaCy search cache. The bigger the swarm the more data is available. It seems you can also initiate your own webcrawling to increase the size of the cache you share. So the searches are completely decentralized and unmodified by any profit-motivated algorithms and come directly from other users searches/crawling. It also seems impossible to be tracked this way. You could see that your IP was connected to the swarm, but it doesn’t seem possible to know what it’s doing on the swarm because there is no central server to log it, just a bunch of direct connections between computers in the swarm.

I don’t know correctly, for example YaCy site describes that.