As elliotcarlson pointed out the issue isn't that answers can be determined automatically, it's that the cost of determining the answers can be amortized over all uses. It's the same vulnerability problem as rainbow tables. With rainbow tables you spend a lot of (automated) effort computing hashes for password guesses, the key advantage of this tactic is that it is widely applicable to every naive use of that hash function.

The amount of effort for a human to go through the list of answers and come up with answers may be non-trivial, but once completed it's applicable to every single use of the plain-text captcha system. That's bad.