> if there are no hits in the dataset showing a trip from the starting location that you know to the ending location that they claimed, then they didn't go wher

> if there are no hits in the dataset showing a trip from the starting location that you know to the ending location that they claimed, then they didn't go where they said they did. As others have pointed out, differential privacy is not obtained just by blurring the points. But I'd like to point out that what you wrote above is closely related to exactly the original definition of DP. DP guarantees that you cannot get a statistically significant difference between a dataset with one element removed (e.g. a trip in your case). That implies that the difference is also insignificant if you add a point. In other words DP addresses exactly your concern here.