C2PA Investigations

Summary

This is the blog version of my talk at the IPTC’s Photo Metadata Conference online conference. Its title is the one the conference organizers slapped on my session without asking; I was initially going to object but then I thought of the big guitar riff in Dire Straits’ Private Investigations and snickered. If you want, instead of reading, to watch me present, that’s on YouTube. Here we go. Hi all, thanks for having me. Today I represent… nobody, officially. I’m not on any of the committees nor am I an employee of any of the providers. But I’m a photographer and software developer and social-media activist and have written a lot about C2PA. So under all those hats this is a subject I care about. Also, I posted this on Twitter back in 2017. I’m not claiming that I was the first with this idea, but I’ve been thinking about the issues for quite a while. Enough self-introduction. Today I’m going to look at C2PA in practice right now in 2025. Then I’m going to talk about what I think it’s is for. Let’s start with a picture. This smaller version doesn’t have C2PA, but if you click on it, the larger version you get does. Photo credit: Rob Pike I should start by saying that a few of the things that I’m going to show you are, umm, broken. But I’m still a C2PA fan. Bear in mind that at this point everything is beta or preview or whatever, at best v1.0. I think we’re in glass-half-full mode. This photo is entirely created and processed by off-the-shelf commercial products and has content credentials, and let me say that I had a freaking hard time finding such a photo. There are very few Content Credentials out there on the Internet. That’s because nearly every online photo is delivered either via social media or by professional publishing software. In both cases, the metadata is routinely stripped, bye-bye C2PA. So one of the big jobs facing us in putting Content Credentials to work is to stop publishers from deleting them. Of course, that’s complicated. Professional publisher...