> due to the actions of... users who were taking advantage of it to try and direct users to potentially dangerous websites.

Ok, the engineer in my is really curious. Apparently, you aren't (too) worried about users including malicious links or content in the original post, but some hole in your security protocol makes it more likely that users can successfully insert malicious contents or links by editing an existing post. The only thing I can imagine is that some automation (or some human) examines the original post for malicious content, but not so much for edits to the post...