Writing in the Age of GenAI
The Language Learning Models of today are capable of doing a lot. How well? That’s what we’re trying to find out here. This blogpost is for the creative writers who want to preserve their voice as they experiment with AI tools and their uses. The purpose is not to find ways for AI to be your process, but to find out if it’s worth incorporating it into your process.
For those who wish to learn more about my approach and attitude towards AI and creative writing, take a look through my introductory post.
Incorporating CoPilot into the Creative Writing Process: Line by Line
CoPilot is . Like ChatGPT, it is a Language Learning Model capable of producing novel responses from a well constructed prompt. For more insight into how to structure a prompt, my previous blogpost outlines a step-by-step process for iterative prompting.
I advocate for an approach where you write freely and use CoPilot at any point of major friction, building on the precedence of successful writers like Vauhini Vara. A line, whether it is a complete sentence or not, is defined here as a complete thought. By going line by line, writers must still put their own thoughts forward, building up on them block by block to ultimately construct their own pieces. CoPilot is then simply a tool to assist the writer along the way.
CoPilot has an exciting feature in its Creative Voice option, which promises to generate more imaginative outputs. CoPilot set to Creative Voice will be the version that will be used for the exercises below.
Exercise 1: Expanding On a Base Line
A common use of Generative AI is unsurprisingly to have them generate new content off a prompt. Here, we looking to see how effective CoPilot is at taking an idea and running with it. In your process, this may be at a point where you know that more can be added but your brain just isn’t . Instead of coming back to it later, perhaps consider seeing what can be done with a chatbot buddy.
Here is a sentence begging to be expanded upon:
“A magnetic mold covers the room, pulsing.”
General guidelines for prompts:
- put the Chatbot into a role (“Act as a science fiction writer”)
- provide a structure for the output (rephrase the following sentence… five different ways )
- be even more specific (“…who is focused on evocative language”, “expanding on it to emphasize the horrific and unknown”)
- provide a line for the Chatbot to work off of (“A magnetic mold covers the room, pulsing.”)
As demonstrated, you can also include multiple parts to the prompt.
I prompted for “evocative language” to test how CoPilot deals with a looser objective. I will also prompt for “adding another clause and an additional sentence of description” to test how CoPilot deals with a more concrete objective.
See the results for yourselves.
Evocative Language:
In trying to rephrase the original sentence, CoPilot’s offerings feel distinct from one another, which is a good sign. The variations in words and descriptions that it uses are nice, although it suffers from being too expository with its added content.
Here, CoPilot’s Creative Voice fails to output as per the prompt’s instructions. There is no additional sentence, although it does offer another clause. The resulting outputs come off as stiff however, even as it offers a similar breadth of interesting words and phrases.
CoPilot has a unique feature in its auto-generated prompts recommendations that ask smart and relevant follow-up questions. This feature has the potential to save a lot of mental space during the writing process and can prove incredibly valuable.
Exercise 2: Modifying Existing Lines
Expansive Rephrasings
There is also the scenario where you want an editor. You already have an idea and you want to make sure that it’s conveyed effectively. It does not have to be a sentence you see as needing work either. I wrote this sentence during a writing activity as year ago, and believe that it stands strong by itself already. However, maybe I want to see if there is another way to phrase it that comes off stronger. The components for success are there, but for better or worse we often are a bit fixed in our styles and preferences. Language Learning Models may prove useful in breaking those subconscious tendencies when the occasion calls for it.
Here is a sentence that could potentially be stronger rephrased:
“There’s restrictions, guidelines painted on the pavement and suspended on wires twenty-two feet in the air that tell us not to get ahead of ourselves, to consider others in our momentum.”
The Guidelines are the same as above. Here I focused on preserving the meaning of the sentence and then on preserving the structure.
Preserving Meaning and Sentiment:
The outputs read well, tactfully reinterpreting different parts of the sentence into different contexts without going overboard. What’s interesting is in a prompt that does not acknowledge preserving the structure, it still maintains the structure of the sentence. As a result, these all feel like the same sentiment and are mildly reworded or have a slightly adjusted context.
Since Copilot was already good about preserving syntax, it was interesting to see how it would interpret a prompt where it was specifically prompted to. Here, it adheres very strictly to the form and phrasing of the original sentence. Notably, the chatbot’s offers a consistent array of strong, evocative words.
Takeaways
CoPilot shows an attention to purpose in including its Creative Voice, and that attention in addition to its other quality of life features puts it as the most effective chatbot for the line-by-line approach to the creative writing process that I’ve come across so far.
The bolding, while potentially distracting to some, does a good job of emphasizing the core ideas of the outputs. It is easier to cherry pick out interesting words and phrases skimming through in this way.
The autogenerated follow-up prompts are excellent in prompting not just the chatbot, but the author in exploring further angles on their pieces.
Curious about how other LLM Chatbots measure up? Consider looking through these other blogposts where I go through the same exercises with the exact same prompts: