Affirming the consequent
The IDC argument can be logically expressed as follows:
• If an object is designed by an intelligence for a purpose, some intelligent being will use it for that purpose.
• There are things that intelligent beings use.
• Therefore, all those things are designed by an intelligence for the purpose.
• If an intelligent being designs an object in order that an intelligent being will use it for a purpose, the design will precede the usage.
• People (intelligent beings) use evolved products.
• Therefore, the process by which those products evolved was intelligent design.
The problem is, IDolators seem to define "design" in bizarre, inconsistent, and inconstant ways. If one defines "design" as utility for an intelligent purpose, the logic above is correct, but we're stuck in the meaningless world of extreme nominalism. As something of a realist, I'm more inclined to think that the concept of "design" has some intrinsic meaning, and that we can't just define it to be anything intelligent beings use for a purpose, or in any other customized way that fits IDeology but not common usage.
The same is true of the first form, but I don't think the temporal aspect is key. Rather, I think the temporal error is different. Dembski seems to be claiming that the process by which an object came to be can become "design" by virtue of its subsequent use. An object is used by an intelligent being and we can now start calling its ontogeny "design."
If we impose a realist perspective, we get to a slightly deeper insight into the leaky ducts of an IDolator's mind. If Dembski has a reasonable understanding of "design" as a real thing, his argument is that anything used by an intelligent being must have been designed, even if no evidence of the process exists.* That argument is very much closer to post hoc ergo propter hoc. It relies on the belief that only things that were intelligently designed can be used for a purpose.
Since that's false, the syllogism collapses as a temporal argument. It would claim that before something could be intelligently used, there must have been a process of intelligent design, and the intelligent design must have preceded the intelligent use. But since a rock can be reconceptualized as well-suited to an intelligent purpose, we know that to be false.
Or consider this. A wrench is intelligently designed to tighten nuts and bolts. I've used a wrench to pound things into place (I knew I shouldn't, but I did anyway). Does my use of a wrench as a mallet mean that the wrench was designed to be used as a mallet? No, of course not. So the argument doesn't even hold for intelligently designed artifacts.
*I'm oversimplifying here. The object has to be complicated, too. But since definitions of "complicated" shift within a single work by Dembski, that's a touchy concept. Plus, I can argue that anything is complicated if I try hard enough. So forget that part of the story, it's a barrier made of thinly sliced Swiss cheese.