Generative AI is increasingly becoming a part of our everyday devices and interactions as it processes and filters information for our consumption. In this process of processing what we perceive, what is lost in the translation?
Persuasion, style, tone and nuance seem to be the casualties. When an AI breaks down a complex topic into its facts, it often strips away the author’s intentional rhetoric, which facilitates comprehension in a way the author intends. There is a homogenising or blanding effect in the process of distilling a piece down to its core argument. We might even reach the point where my AI will help write something in a way that only I can write, and your AI will interpret it for you in a way that you like to consume it.
The utility of AI is undeniable. It can summarise the key takeaways from a complex report or get the gist of a rambling email. It can explain quantum mechanics in a language that you can understand. It can sift through legalese and identify potential concerns in a legal document. It can summarise the day’s news so you can stay on top of developments. All of these are great and increase our productivity significantly. But there is a hidden cost to this frictionless clarity.
Being mindful of our information diet is essential. This might mean that we need to consciously curate the information we consume and engage with the author’s original writing to build familiarity with their work and style. AI can aid in reading iteratively by using an AI summary to skim a complex piece before diving into it for a more detailed and nuanced understanding. Perhaps even consciously planning for serendipity can lead to a meaningful encounter with an idea outside our typical sphere. And for authors, it means making your writing in a way that is more engaging than an AI-generated summary.