OpenGraph allows to get a preview of the content.
Allowing access to that preview is implicit and it looks fair to access the image from the media site.
However, republishing that preview elsewhere is still publishing and the author of the post seems to have missed that. Instead he should just publish the link to the media site and let the client that browse his archive access (download) the OpenGraph preview by itself.
Twitter/X does that republishing, but having the license for that republishing is their problem.
I'm curious to know if that preview was part of the Twitter archive. Because it doesn't qualify as "your" content.
A technical solution would be to implement the OpenGraph preview on the client side (JS), instead of having the preview in the archive and hosted/served by it.
I am unsure if this really would work legally, let alone stopping a licensor from demanding money from you in the first place.
Does is really matter where the image is served from? If its displayed on your website, does it really matter if its client side JS or static HTML that renders the img tag? I think neither does it matter if you load the image from your own server or from a remote (the original OpenGraph source).
* it would at least help to go under the radar of some scrappers which don't run JS.
* not serving the infringing content might give some more weight for defense of a fair use position before a court (but IANAL)
Not easy to implement because of CORS limits. The usual limitation of CORS is bypassed by implementing a proxy, but that's exactly the kind of solution to avoid here as the proxy would be the licensor's target.
However, republishing that preview elsewhere is still publishing and the author of the post seems to have missed that. Instead he should just publish the link to the media site and let the client that browse his archive access (download) the OpenGraph preview by itself.
Twitter/X does that republishing, but having the license for that republishing is their problem.
I'm curious to know if that preview was part of the Twitter archive. Because it doesn't qualify as "your" content.