Known threat vs. unknown threat

Cybersecurity techniques depend on the identification of "static" known threats, such as viruses, rootkits, Trojans, and botnets. It is straightforward to identify and scan for this type of threat with automated software by matching the malicious code to a signature in a database of known malware.

An example of a known unknown is that malware authors can use various obfuscation techniques to circumvent signature matching. The exact form that such malware will take is unknown, but its likely to use and operation within an attack is predictable.

Recycled threats - combining and modifying parts of existing exploit code to create new threats that are not as easily identified by automated scanning.

Unknown unknowns - completely new attack vectors and exploits. One of the purposes of security research is to try to discover these, using techniques such as analysis of data collected in honeypots and monitoring of discussion boards used by threat actors.