I'm interested in a comprehensive set of related notions of authenticity of digital objects in the context of digital preservation. In this sense, I would be curious to know how we should be thinking about authenticity in a forensic sense, a diplomatics sense, an archaeological sense, in terms of bit level integrity, and whatever other senses that those working on digital preservation issues feel it is necessary for us to attend to.
I am an archivist and have been interested and thinking about authenticity for a long time ;-) especially since I started working in digital preservation. Authenticity and how to ensure it is maintained across time is a crucial question for the archivist and not just for those who deal with digital records.
The definition given in a previous reply here is the basic one most archivists use but to expand it somewhat I think authenticity involves other record characteristics of reliability (ie. it is what it says it is), integrity (ie. it is complete and unaltered) and useability (ie. the record can be retrieved and understood). Assuring these things means that the record is authentic and will remain so. Note that this does not mean the digital record must always be maintained in the original format in which it was ingested into a digital archive, and it is particularly important to note that authenticity is not saying anything about the truthfulness, trustworthiness and veracity of the contents of the record, only that it is what is was said by its creators to be and was created by them in the context and at the time they say it was.
There was some a lot of good archivally-based work done on authenticity and digital records by the InterPARES project at the University of British Columbia (http://www.interpares.org/) and the project published a lot of documents about the outcomes of the research. Look particularly at InterPARES 3 which was specifically addressing digital preservation and authenticity.
In the context of digital preservation ensuring authenticity is not straightforward. In my experience it involves making sure all the recordkeeping and contextual (provenance etc) metadata is tied in some way to the record for as long as it is in the archival system (can be part of the record, can be in a database but permanently linked to the record), ensuring that there are rigorous controls on access to the records, ensuring that there are no unauthorised or unseen changes to the bit level integrity of the record (its fixity), and, critically, keeping a record of everything that is done to the record as it journeys through the OAIS.
I am not an archivist, but the definition for authenticity you often hear from archivists is the following:
"An authentic record is one that can be proven to be what it purports to be, to have been created or sent by the person purported to have created or sent them, and to have been created or sent at the time purported".
(Also quoted in http://www.ijdc.net/index.php/ijdc/article/download/113/100, that is also a good discussion of misinterpretations of this concept by one of the most reliable professionals in this field).
In practical terms, my understanding is that (and I would like to be corrected in case) the archive acquires at deposit time the available evidence of what the object that is being accessioned - but it does not try to establish its authenticity in an archaeological / forensic sense. Authenticity in archival terms means that the archive is responsible for the object still remaining "what it purports to be" AFTER it has entered the archive - if it was a fake to start with, this is not the archive's problem, but it needs to be protected from falsifications of any kind while it is within the archive itself.
Of course digital archivists I have worked with have tried to transfer this notion to digital objects, this is just an example but a good one because it builds on OAIS:
In a general way, I find very interesting for this discussion the notion of artifactual value:
I will post again if I find anything more conclusive.