They are different because of aggregation: the hundreds of signals an organization has from your activities are combined into thousands of derived signals, whic

They are different because of aggregation: the hundreds of signals an organization has from your activities are combined into thousands of derived signals, which you probably would not have offered the organization freely. Just for example, by combining several data points, Skynet can infer that those people are acquainted and add that fact to the corpus about you both. You would never give a retailer a full dump of everyone in your address book, but it's pretty much given they know it.