Ah yes, now that you note it I remember this plan from last year. Thank you.

ZimmerBob: The plan to run the unmanned Dragon 2 mission to ISS preceding the in-flight abort test was announced shortly after the pad abort test last year, IIRC. The idea is that the Dragon 2 that’s going to be used for that 1st unmanned ISS rendezvous is going to be pretty much identical to the unit that actually delivers crew to ISS a few months later. SpaceX intends to fish that first ISS-visiting Dragon 2 out of the drink, clean it up and use it for the in-flight abort test. The reason SpaceX gave at the time for wanting to do things this way was that the Dragon 2 that makes the first unmanned ISS visit will incorporate all the operational subsystems and engineering changes of the initially operational manned version. Re-using it for an in-flight abort test will both demonstrate Dragon 2 reusability and provide maximum fidelity so far as how a real ascent emergency would play out compared to using the same earlier-stage-of-development vehicle from the pad abort test. Sounded reasonable to me at the time and still does.

When SpaceX first started the ocean landings (no barge, just trying to get it vertical as it reached the water), NASA was very interested in the tests and monitored them with SpaceX approval. SpaceX got something out of it, too, I don’t quite remember, but perhaps data that NASA collected. I don’t remember them mentioning at the time that NASA was intending to use the information gathered for Martian EDL of heavy payloads.