Here is a link to the European Commission for the Environment's website for the revision of the directive:
Animals in science
I now there isn't a systemic problem, Thales, and I strongly doubt there are any cases of ceph abuse in research labs and aquaria (other than possibly the one that started this thread, which is in, in fact, an AZA member aquarium), but consider the impact of welfare laws for vertebrates; for example, the requirement to establish humane endpoints results in considerable reduction of suffering because it establishes points at which an animal must be removed from an experiment, either because the experiment has ended or because the animal is sick or hurt. There is no such requirement for invertebrates (although I know of at least one case in the context of pain research, on tonmo user robyn's blog -
http://www.tonmo.com/blog/entry.php?66-Getting-published.... - , where research has not been allowed to go forward because of animal welfare concerns about invertebrates. I suppose institutional review boards may cover some of this in the research setting).
The AZA site doesn't have much for nonmembers other than a section on conservation initiatives (e.g. the white-spotted octopus). I'd guess you have an account that you can use to access information for members, and if you can find any, please share it with us if possible.
I guess I'm approaching this from this viewpoint: the common octopus, at least (best-studied among the cephalopods), is rightly considered an honorary vertebrate in the EU for good reasons.
First, it has a brain-to-body mass comparable to many birds and mammals.
(I was actually present, in fact, for an informal brain-to-body-mass determination of a Humboldt squid in October at the National Science and Engineering Festival; I was working with Dr. William Gilly and he was doing some stuff for a kid whose birthday it was that day. Using a small luggage scale, we determined that it had a brain-to-body-mass ratio somewhere between "a frog and a dog". This is a Humboldt squid, not a common octopus; the Humboldt squid is considerably less behaviorally complex, as far as I'm aware, than a common octopus!)
The brain-to-body-mass ratio (or encephalization quotient) and other measures such as the surface area to volume ratio of the brain are good rough estimates (not perfect) of the complexity of an animal's cognition.
Here is a list of relevant publications:
Google Scholar
I shouldn't have to go far into behavioral complexity to explain why this probably would justify some sort of legally-codified protection of cephalopods, because I'm assuming most people on this forum know about it.
In addition, again referencing Robyn's blog post, there is some evidence that the peripheral nervous system of octopuses may take care of at least some amount of sensory and motor processing, but nobody seems to know how much and whether the CNS is involved in processing pain.
In general, at least for me (and I'm going to carry out research on cephalopods for a living, so I know I'll be dealing with this, too), it seems kind of selfish to give a higher priority to bureaucratic concerns such as mountains of paperwork, rules that might 'backfire' (can you provide instances of when these sorts of things backfired?), advance permissions, or some claim that these directives get in the way of doing science than to the well-being of our captive cephalopods. These sort of claims - and I sincerely don't wish to offend, but this is quite true - are old hat when it comes to vertebrate welfare.
You may not abuse the cephalopods in your care, even unintentionally, but perhaps somewhere along the line there may be some idiot who does.