There are times when you can get away with using a delayed start date for a cal interval. However, generally, in an audit, you would have the burden of proof to demonstrate to the auditor why during the intervening time the accuracy would not have degraded in the instrument.
Items which change value strictly due to use versus passage of time, that may be acceptable. In a previous employer, we used high temperature thermocouples which have a life span of about 6 months when used at around 1000 to 1200 Degrees Celsius. However, that aging was due to the high temperature. So we had a document in our quality system stating that our procedure was to place them in environmentally controlled storage. Then, when they were given to the user, the calibration interval began. This could have been true for items which only age/drift/change during use, and definitely do not change in controlled storage.
But for everything else, the cal date is when it was actually calibrated.
My recommendation would be as in the previous post: to either live with the existing cal date/due date, or get them re certified.
As to the validity of the OEM certificate, some are good and some are not. I am currently in a high volume lab with pretty strict guidelines in that regard. We review every one of the OEM certificates on new instruments. Quite often they are fine (from the large, reputable OEMS (such as Fluke, Tektronix, Agilent, Mitutoyo, etc.). But some are not. I would weigh the validity of the particular certificate as to whether to accept it.
Sorry if any of this is bad news. Just have to shoot as straight as I can, and try to help.