I was wrong — it's not the [Harvard multimedia video](http://multimedia.mcb.harvard.edu/media.html). It's an independently generated copy. I grabbed a few images from the DVD I got at my [truncated visit to the _Expelled_ screening](http://scienceblogs.com/pharyngula/2008/03/expelled.php), and here, for instance, is the segment that shows that striking kinesin motor protein towing a vesicle down a microtubule. This is the version in the _Expelled_ movie:
Now here's an equivalent frame from the actual Harvard video.
Now I'm embarrassed to have mistaken one for the other, since the _Expelled_ version is of _much_ lower resolution and quality. However, do notice that they both have roughly the same layout and the same elements in view; this is a remarkable, umm, coincidence, since these are highly edited, selected renderings, with many molecules omitted … and curiously, they've both left out the same things.
Another curious coincidence: you've heard of the concept of [plagiarized errors](http://www.talkorigins.org/faqs/molgen/), the idea that the real tell-tale of a copy is when it's the mistakes that are duplicated, in addition to the accuracies. In this case, I [previously criticized the Harvard video](http://scienceblogs.com/pharyngula/2008/02/buffeted_by_the_winds_of_chanc.php) for a shortcut. That kinesin molecule is illustrated showing a stately march, step by step, straight down the microtubule. Observations of kinesin show it's more complex, jittering back and forth and advancing stochastically. That's a simplification in the Harvard video that is also present in _Expelled_'s version.
It's clear that what they did was brainlessly copy what they saw in the original. I don't know whether this is actionable anymore — that they slapped together a look-alike video to cover their butts makes the issue much more complicated.