Your use of the word "innate" at the end of the question itself looks to me like an example. When speaking literally, "innate" means that is something that a person is born with, and not something that they learn. Linguists are pretty familiar with the term thanks to Chomsky's proposal that humans are genetically endowed, prior to experience, with specific knowledge of the nature of languages. I assume that you recognise that your use of "innate" in saying that "words have an innate literal meaning" would mean that humans already know words and meanings for (all) human languages at birth. Since you surely don't believe that, then you would be using the word "innate" with a different meaning. Some people occasionally use "innate" to mean simply "deep-seated; automatic". It is not hard to see how such figurative usage arises: if some cognitive faculty is actually innate, then it will probably be deep-seated and automatic.
At some point in the historical development of a language, a word such as "innate" which is frequently used figuratively in a particular way becomes polysemous. An example of that change is the word "cow", which historically refers to female Bos taurus, but is now applied to female camels, bison, alligators, komodo dragons, dolphins, eland, elephants among others. Another more recent example is the word "virus", which used to refer only to a specific tiny kind of pathogen, but a few years ago was figuratively extended to a kind of computer program. This extension has been so thorough that I'm betting that in 20 more years, almost nobody will know that the term used to refer just to biological viruses.
It is extremely difficult to know whether the meanings associated with a particular word are literal or figurative, because "meaning" is understood to be both a social construct and a fact of individual psychology. You learn social facts about meaning the same way you learn any other social facts -- you observe it as used by others, and you decide (somehow, the how being a big mystery) what the essential defining characteristic of that fact is.
I disagree with Colin Fine's apparent assessment of the importance of literal meaning as being useful to just a select few people. Any person who enters into a legally binding agreement, or is subject to any laws, will need to be able to discern literal meanings, if in that society laws and agreements are interpreted according to meaning. These days, that is pretty much everybody.