Most scientists get lots of mail from dubious companies with invitations to publish. Unsubscribe buttons don't work, or simply don't exist. This led one computer scientist to respond to the mail with submission of a paper called "Get me off Your Fucking Mailing List", consisting of pages full of this same sentence. It got accepted.
As hilarious as this is, it indicates a big problem. Open access journals have advantages compared to traditional big publishers, which I've discussed before, but its freedoms can also facilitate predatory publishing. These companies ask authors for fees for reviewing and editing, without actually performing these services. They hide their publishing fees, or continue to increase their fees after submission, while not allowing the authors to withdraw their manuscript, making it impossible for them to publish anywhere else.
Bogus journals might steal the names of respected academics and editors to make their business look real, although that doesn't mean necessarily that they were never in touch. Here's the account of a scientist who joined the editing board of a new journal, but when he and his colleagues tried to withdraw, they got no response and their online biographies were not taken down. If you're looking for more examples, look for the complaint the FTC (U.S. Federal Trade Commission) filed against OMICS and others in 2016, which seemed to halt their business, though they still exist today.
These companies can appear very large, and therefore trustworthy (the OMICS International group owns and operates 700 journals). They might claim to have a lot of partners in academia, and go so far as to falsely advertise conferences, charging attendees over $1000, while the people on the speaker list actually know nothing about the event. They can lie about the platforms (such as PubMed) on which they are indexed, and claim a false impact factor. In short, anything goes to take your money.
Because predatory publishers label themselves as open access, the two might seem connected, but the rotten apples do not take away impact or validity from real open access publishers. However, this distinction proves hard to make on the face of it, even for governments. Authorities or scientists might not recognize a business to be illegitimate until it's too late. The publish-or-perish culture in academia amplifies this, especially in countries where you need to publish in foreign papers to be recognized in your work field. This last point contributes to India and Pakistan being popular for predatory publishers, although the U.S. is not far behind.
Here's an excellent lecture from the University of Virginia about predatory publishers and how to avoid them. They advise scientists to confirm the publisher's reputation with the thinkchecksubmit.org checklist, the COPE's Directory of Open Access journals (DOAJ), the Open Access Scholarly Publishers Association, and Publons' list.
Check your publisher - keep your manuscript safe.