Wittgenstein introduces the concept of language games to account for the flexibility of language across contexts (Wittgenstein, 2009). Depending on the context, a word has multiple meanings. For example, the word “slab” has different meanings to a builder and a butcher. To further this point Wittgenstein argues “don’t think, look”. This argument illustrates the need to attend to specific cases as opposed to always relying upon generalizations.
A generalization will impose the meaning of a word upon reality, leaving the nuance of the case ignored. Since the objects we encounter in the world have multiple aspects, if we think instead of look, we will fail to notice beyond a single aspect (Wittgenstein, 2009). To illustrate, fire can both be devastating and lifesaving, being able to recognize both aspects of fire will allow us to run from danger and move toward survival depending on the context.
This brings us to the Generative AI Chat GPT. We can ask the AI questions and it retrieves web data relevant to the question and presents an answer to us as a narrative. The answer contains a critical element, presenting pros and cons. It is reasonable to conclude that we have received a balanced answer, making further exploration unnecessary.
However, as we argued before on this blog, the answer lacks perspective. Since the AI is not human, the answer cannot relate to the context with which we are concerned. It is up to the human user to take the AI generated answer for what it is worth and fit it to the case.
Wittgenstein’s point “don’t think, look” can be applied to working with Chat GPT. The AI will provide a generalization, and then the user should nuance the generalization to the case. If the generalization is “forced” into the case, then this impairs contact with reality, the user has ceased to look. Generative AI should provide stopping points on a journey of discovery, it should not stop the user noticing and attending to the world around them (Mcgilchrst, 2021). It is our role as human beings to provide the perspective which AI lacks..
Reading
Wittgenstein, L. (2009). Philosophical Investigations. John Wiley & Sons.
McGilchrist, I., 2021. The Matter with Things: Our Brains, Our Delusions, and the Unmaking of the World. What Then is True?. Volume Two. Perspectiva Press
