11 writing aids that will be replaced by ChatGPT

Reading Time: 27 minutes

When a disruptive new technology comes along, it often leaves the corpses of other devices in its wake. The smartphone, for example, replaced or is in the process of replacing standalone GPS devices, radios, alarm clocks, cameras, scanners, audio recorders, video cameras, flashlights, dice, notepads, hand-held games, calculators, landline phones, physical books, newspapers, paper maps, credit card terminals, encyclopedias, dictionaries and thesauruses, compasses, walkie-talkies, Rolodexes, watches, payphones, fax machines, film projectors, telegrams, and hand-held translators.

I didn’t come up with that whole list — ChatGPT suggested the last eight items after I’d run out of ideas.

The same thing is happening now with generative AI. ChatGPT and other large language models that are about to follow in its footsteps are a new evolutionary step in technology.

A very big step.

The more I’ve been playing around with the system since it was first released on Nov. 30, the more impressed I am with its potential — and the more concerned I am about the future of the writing profession, and, more selfishly, my own role in it.

But today, I’m going to be focusing on the potential. Specifically, the way ChatGPT eliminates many of the tools I’ve got bookmarked and reference books I’ve got sitting on my shelves.

(Image by Maria Korolov via Midjourney.)

Thesaurus

My favorite online thesaurus is WordHippo. But ChatGPT can be a lot more context-sensitive.

Say, for example, I want an alternative to the word “chair” that sounds more futuristic.

I gave it this prompt:

can you give me some futuristic or sci-fi alternatives for the word “chair” or “seat”?

ChatGPT:

Sure! Here are a few more alternatives to the words “chair” and “seat” that might be used to describe futuristic or sci-fi seating:

    1. Accelerator chair: a chair that provides high-speed acceleration or movement, often used in transportation or simulation.
    2. Antigravity chair: a chair that allows the user to float or levitate, potentially using advanced technology such as a gravity field generator or anti-gravity force.
    3. Cryogenic chair: a chair that uses cryogenic technology to preserve or revive the user, potentially in a state of suspended animation or deep freeze.
    4. Cyborg chair: a chair that enhances the user’s physical or mental abilities, using cybernetic implants or interfaces.
    5. Holographic chair: a chair that projects a holographic image of the user or environment, potentially using advanced technology such as lasers or volumetric displays.
    6. Neural chair: a chair that connects directly to the user’s brain, potentially using advanced technology such as brain-computer interfaces or neural implants.
    7. Telekinetic chair: a chair that allows the user to move or manipulate objects using their thoughts or mental powers, potentially using advanced technology such as telekinetic fields or psychic amplifiers.

I hope these suggestions are helpful! Let me know if you have any other questions.

Those were okay, but I wanted something that didn’t have the word “chair” in it.

Prompt:

give me some that don’t have the word “chair” in them

ChatGPT:

Sure! Here are a few more alternatives to the words “chair” and “seat” that might be used to describe futuristic or sci-fi seating, and that don’t include the word “chair” in them:

    1. Gravity lounger: a comfortable, reclining seat that adjusts to the user’s body and movements, using sensors and motors to provide a relaxing and ergonomic experience.
    2. Teleportation pod: a small, enclosed compartment or seat that allows the user to instantly travel from one location to another, potentially using advanced technology such as teleportation or wormholes.
    3. Virtual reality couch: a comfortable, reclining seat that immerses the user in a simulated environment, using virtual reality technology such as headsets or haptic feedback.
    4. Augmented reality couch: a comfortable, reclining seat that enhances the user’s real-world environment with digital information or graphics, using augmented reality technology such as smart glasses or projection screens.
    5. Antigravity recliner: a comfortable, reclining seat that allows the user to float or levitate, potentially using advanced technology such as a gravity field generator or anti-gravity force.
    6. Cryogenic capsule: a small, enclosed compartment or seat that uses cryogenic technology to preserve or revive the user, potentially in a state of suspended animation or deep freeze.
    7. Neural couch: a comfortable, reclining seat that connects directly to the user’s brain, potentially using advanced technology such as brain-computer interfaces or neural implants.
    8. Telekinetic throne: a grand and imposing seat that allows the user to move or manipulate objects using their thoughts or mental powers, potentially using advanced technology such as telekinetic fields or psychic amplifiers.

I hope these suggestions are helpful! Let me know if you have any other questions.

But these are all two-word phrases. I want something short and simple.

And it spit out a few more ideas — a hoverseat, an accelerator, a pod, a cryopod, a teleporter…

You can see where I’m going with this.

You don’t just give it one word and get back a list of synonyms — you can explain what you want the synonyms to do, what your purpose is, and it will try to find the words that fit that purpose.

For example, if you’re writing fantasy instead of sci-fi, you can ask for fantasy words, or words appropriate to the medieval period, or words with foreign influences, or made up words that sound like something else.

(Image by Maria Korolov via Midjourney.)

Name generators

Some characters come to me fully-formed, names and all. More often, I don’t have time to come up with new names for every supporting character, city, street, shop, business, or tavern that my story mentions.

I turn to online name generators for ideas. My favorite is FantasyNameGenerators.com, a site with hundreds of generators for everything from elvish names to business names to alien races and planets.

I use the site so often that I have it set up to open automatically whenever I start my day’s writing.

I don’t know what magic they’ve got happening behind the scenes, but whatever it is, it works. For example, when I look for troll names, it gives me these ten random suggestions: Zulrajas, Vuzembi, Vekuzz, Mohanlal, Zebajin, Hoodah, Vulzal, Zelaji, Tazingo, and Zebajin. If I don’t like any of them, I can hit a button and get a new set. I love this tool.

Now, let’s try ChatGPT.

When I say, “give me some troll names,” it responds with:  Trolldalf the Grey, Grumplefist, Snarltooth, Razzlefrazz, Gorgon the Troll, Manglehorn, Sludgespit, Gnashfang, Tanglemane, and Miregut.

If I don’t like any of them, I can rerun the same query, or ask it to make the names scarier, or funnier, or longer, or shorter, that sound vaguely Greek, or include some weapons in them.

Here’s what it gave me when I asked the names to be alliterative, and include weapons: Axe-wielding Axgar, Bow-hunting Borgan, Claw-clutching Clagor, Dagger-dealing Dargar, Hammer-hurling Hargar, Mace-swinging Margar, Spear-throwing Sargar, Sword-swinging Sorgar, Polearm-poking Porgar, and Crossbow-cranking Corgar.

I don’t know about you, but I’m impressed.

(Image by Maria Korolov via Midjourney.)

Character generators

Character generators are online tools that create random characters complete with backstories.

I like Character Generator UK and RanGen’s Quick Character Generator. They’re particularly good for generating minor characters where you don’t care too much about the details. They’re also good as placeholders, so you don’t have to interrupt the flow of your writing — you can always come back and flesh out the characters later.

ChatGPT can take this to another level.

Prompt:

Create a character outline for a gardener named “George Pen” who lives in an English village in the 1940s. He’s an eccentric supporting character in a cozy murder mystery.

After a little back-and-forth, here is a snippet of what ChatGPT came up with — the actual list is quite a bit longer, and I cut it down because really, how much do you want to read about George?

ChatGPT:

Here is a summary of everything we know about George Pen so far:

Biography:

    • George Pen is a middle-aged man who was born and raised in the small English village where he still lives.
    • He is an only child, and his parents were both avid gardeners who instilled a love of horticulture in him from a young age.
    • He is well-known and well-respected in the village for his expert gardening skills and his kind, friendly demeanor.

Appearance:

    • He has salt-and-pepper hair that is starting to thin on top, and he often has a bit of dirt or mud smudged on his face or hands from his work in the garden.
    • George dresses simply, usually wearing a pair of well-worn overalls over a tweed jacket or a button-down shirt with the sleeves rolled up, paired with a pair of sturdy gardening boots.
    • He values comfort and utility over fashion, and is rarely seen in anything other than his gardening attire.

Hobbies:

    • George is an avid gardener, and can often be found tending to his sprawling garden or experimenting with new plants and techniques.
    • George is a collector of rare or unusual items, and has a small museum in his garden where he displays his collection. His collection might include antique gardening tools, unusual plant specimens, or other curiosities related to his love of horticulture. He might also enjoy sharing his knowledge and expertise with others, and might be happy to give tours of his collection to anyone who is interested.

Once you have a character description, you can then take it to an image-generating tool like Midjourney and have it generate some character concept art for you. For example, I asked ChatGPT to give me a list of words that describe George Pen and his appearance.

ChatGPT:

kind, friendly, well-respected, passionate, expert, skilled, dedicated, average, middle-aged, average height, average build, thinning, salt-and-pepper hair, smudged face, smudged hands, casual, overalls, gardening boots, clean, well-groomed, avid, natural, sharp, curious, well-rounded, wide-ranging

I put that into Midjourney, adding “portrait of middle-aged man” to the list.

Here are some of character portraits I got in return:

(Image by Maria Korolov via Midjourney.)

I like the one on the top right.

If you’re a writer who likes to have pictures of your characters in front of you as you write,  this is a great way to get character concept art quickly and easily.

(Image by Maria Korolov via Midjourney.)

Plot generators

I rarely get stuck with my plot when writing. When I do, I usually get in my car or go for a walk, turn on my Otter.AI dictation app, and start brainstorming. If there’s no time for that, I might use an online idea generator, like Plot Generator UK or Reedsy’s Plot Generator.  

But they’ve got nothing on ChatGPT.

Say your fantasy detective needs to find a clue to break the case wide open. Someone killed a troll, but left some evidence behind. What could that evidence be?

At first pass, it suggested DNA evidence or surveillance footage. But I’m in a fantasy setting, so those aren’t appropriate.

Prompt:

This story is in a medieval fantasy setting. Give some ideas for evidence that would be useful in the absence of modern technology.

ChatGPT:

In a medieval fantasy setting, there may not be access to modern forensic techniques like DNA analysis and fingerprinting, so the detective will have to rely on more traditional methods of investigation. Here are some ideas for evidence that could be useful in this type of setting:

    1. Physical evidence: This could include anything found at the crime scene, such as a weapon or other object that could have been used in the crime.
    2. Eyewitness testimony: If anyone saw the crime being committed or saw the killer around the time of the crime, their testimony could be useful in identifying the perpetrator.
    3. Alibis: If the detective can establish where the suspects were at the time of the crime, this could help rule out certain individuals as suspects.
    4. Motive: Understanding the motive for the crime could be key in solving the case. Was the troll killed for personal reasons, or was there a more sinister motive at play?
    5. Forensic examination of the body: In the absence of modern forensic techniques, the detective might be able to glean clues about the killer by examining the body and the crime scene. For example, the detective might be able to determine the type of weapon used in the crime or other details about the killer’s appearance or behavior.
    6. Physical evidence from the killer: If the killer left behind any personal items at the crime scene, such as clothing or other belongings, these could provide clues about their identity.

I like that last one — physical evidence from the killer. Let’s see if ChatGPT can come up with any examples.

I wrote: “I like number 6. Can you give me a few suggestions?” It gave me a few obvious ones — clothing or other personal items, weapons, notes or other written materials, personal belongings. How about something more unusual? It suggested religious or magical symbols, unusual art work, or a coded message.

If I want more ideas, I just type “continue” or hit the “regenerate responses” button. If it starts repeating itself, I’ll rephrase the prompt. For example, I might ask it for clues that are magical in nature, or clues that point to other trolls. For example, one suggestion ChatGPT offered was: “A magical weapon: If the killer used a magical weapon to commit the crime, the detective might be able to trace the weapon back to the killer or identify its unique magical properties.”

That’s pretty good.

(AI image generated by Midjourney.)

Outlining tools

Once I’ve got some ideas for what I’m going to write about, I’ll often draft a quick outline to keep me on track.

I’ve used a variety of outlining tools for this in the past. Usually, I’ll do a simple outline in my word processor or Google Docs. Sometimes, I’ll use a spreadsheet like Google Sheets or an online database like Notion. There are also some mind map tools online — I’ve written up a list of 15 free ones here.

But ChatGPT can just create the whole outline for me.

Prompt:

Write me an outline of a scene in which my fantasy detective discovers that a killer used a magical weapon to kill a troll. The killer is another troll. The scene should explain how the detective was able to identify the weapon’s unique magical properties and trace it back to the killer.

ChatGPT:

    1. The detective is called to the scene of a crime where a troll has been murdered.
    2. Upon examining the body, the detective notices unusual wounds on the troll that suggest the use of a magical weapon.
    3. The detective begins to search for clues and eventually finds a small, ornately decorated knife at the crime scene.
    4. The detective recognizes the knife as a rare and powerful magical weapon, known for its ability to slay even the strongest of trolls.
    5. The detective decides to consult with a local expert on magical artifacts to learn more about the knife.
    6. The expert confirms the detective’s suspicions and reveals that only a few of these knives were ever created, and each one is unique.
    7. The detective then begins to investigate any known owners of the knife and eventually discovers that a rival troll was in possession of one.
    8. The detective confronts the suspect troll and, upon further investigation, uncovers evidence linking the troll to the murder.
    9. The detective is able to use the unique properties of the magical weapon to build a strong case against the killer, leading to their arrest and eventual conviction.

That’s not the outline of a scene — that’s the whole book! All it’s missing is a romantic subplot. Which gives me an idea.

Prompt:

Add a romantic subplot.

ChatGPT:

  1. The detective is called to the scene of a crime where a troll has been murdered.
  2. Upon examining the body, the detective notices unusual wounds on the troll that suggest the use of a magical weapon.
  3. The detective begins to search for clues and eventually finds a small, ornately decorated knife at the crime scene.
  4. The detective recognizes the knife as a rare and powerful magical weapon, known for its ability to slay even the strongest of trolls.
  5. As the detective investigates the crime, they are assisted by a charming and intelligent elf, who is also an expert on magical artifacts.
  6. The detective and the elf develop a romantic connection as they work together to solve the case.
  7. The detective decides to consult with the elf about the knife, hoping to learn more about its unique magical properties.
  8. The elf confirms the detective’s suspicions and reveals that only a few of these knives were ever created, and each one is unique.
  9. The detective then begins to investigate any known owners of the knife and eventually discovers that a rival troll was in possession of one.
  10. The detective confronts the suspect troll and, upon further investigation with the help of the elf, uncovers evidence linking the troll to the murder.
  11. The detective is able to use the unique properties of the magical weapon to build a strong case against the killer, leading to their arrest and eventual conviction.
  12. As the case comes to a close, the detective and the elf decide to pursue a romantic relationship, having grown close during the investigation.

Now what the story is missing is some personal stakes for the detective, and a ticking clock. And some red herrings. Each time I ask for more stuff, ChatGPT updates the outline, fleshing it out further.

I don’t know about you, but to me this puts every single other outline tool to shame. And as you write your story, if you decide to move things in a different direction, ChatGPT can generate a new outline for you — just like your phone’s GPS can automatically recalculate your route when you leave the suggested path.

Another way to use ChatGPT for outlining or drafting is to take the transcript you got from your Otter AI dictation, or random notes you jotted down over time, and ask ChatGPT to turn it all into a story outline for you.

Say, for example, I’ve got the following in my notes:

  • There’s an elf with a perky nose.
  • Oh, she an expert on magical forensics. Maybe she teaches at the local community college?
  • The detective normally follows cheating trolls around. This troll’s wife hired her to follow her husband and the husband is killed while the detective is following him.
  • Oh no — the wife is the prime suspect! After all, she knows the guy was cheating on her. and people think that she hired the detective to off him!
  • Everyone is surprised that the detective was able to kill the troll — much less killing him without leaving a trace.
  • She did the king a favor once, though, and he decides to give her a chance to clear her name before throwing her into the dungeon and trying her for murder.
  • The elf helps the detective figure out what magic was used to kill the troll.
  • As they get closer to the real killer, the killer decides that he’s going to kill the elf.
  • I need to figure out some motives for the real killer here. Also a few red herrings.
  • I need to figure out why the detective and elf fall in love, too, and weave it into the plot.

I cut and paste the notes into ChatGPT and ask it to turn them into a fully-fleshed-out outline for a novel. I specify that I want to start right out with the murder, and that I want twists and turns and false leads and a romantic subplot.

On the first pass, ChatGPT misunderstood — it thought that my detective and the elf forensics expert were the same person. I had to tell it that this wasn’t the case and to revise the outline.

Here’s what it came up with:

Chapter 1: The murder of the cheating troll occurs in the opening scene, while the detective is following him.

Chapter 2: The detective is hired by the troll’s wife, who is now the prime suspect in the murder.

Chapter 3: The detective, with the help of the elf expert on magical forensics, begins to investigate the murder and gather clues. The king, who owes the detective a favor, gives her a limited amount of time to solve the case before she is thrown in jail.

Chapter 4: The detective and the elf discover that the troll was killed using a rare and powerful magical weapon, which complicates the investigation.

Chapter 5: The detective and the elf follow a lead that takes them to a rival of the troll’s, who seems like a strong suspect.

Chapter 6: As they continue to investigate, the detective and the elf uncover more false leads and red herrings.

Chapter 7: The detective and the elf pursue a new suspect, who has a motive for the murder but lacks an alibi.

Chapter 8: The detective and the elf discover evidence that points them towards yet another suspect, who seems to have the means and opportunity to commit the crime.

Chapter 9: The detective and the elf follow a trail of clues that leads them to their final suspect, who becomes the focus of their investigation.

Chapter 10: The new suspect reveals their motive for the murder – they were jealous of the troll’s success and wanted to eliminate the competition.

Chapter 11: The killer, realizing that the detective and the elf are getting closer to the truth, attempts to eliminate them.

Chapter 12: In a dramatic showdown, the detective and the elf confront the killer and bring them to justice. The detective and the elf, who have fallen in love during the course of the investigation, decide to be together as the mystery is finally solved and the killer is brought to justice.

It still needs some tweaking and fleshing out. But ChatGPT can do that for you, too.

(Image by Maria Korolov via Midjourney.)

Google and Wikipedia

I use Google quite a bit when writing.

For example, I have a scene where my main character needs to borrow a farm truck from her gardener, and the story is set in a small village in 1940s England. I spent a couple of hours of Googling before I finally threw in the towel.

ChatGPT described the requirements of such a truck for me in seconds. I followed up by asking it to suggest specific models.

ChatGPT:

Here are some examples of makes and models of pickup trucks that a gardener in a small village in 1940s England might have used:

    1. Ford Model T Runabout: The Ford Model T Runabout was a small pickup truck that was produced in the early 1900s. It was a popular choice for small businesses, including gardeners, due to its low cost and versatility.
    2. Morris Commercial J-Type: The Morris Commercial J-Type was a small pickup truck that was produced in the 1930s and 1940s. It was known for its reliability and was often used by small businesses, including gardeners.
    3. Chevrolet AK Series: The Chevrolet AK Series was a line of pickup trucks that was produced in the late 1930s and early 1940s. It was known for its durability and was often used by small businesses, including gardeners, for hauling and towing.
    4. Dodge WC Series: The Dodge WC Series was a line of pickup trucks that was produced in the 1930s and 1940s. It was known for its toughness and was often used by small businesses, including gardeners, for hauling and towing.

That’s way more useful than anything Google could give me, and much faster, too. I know because I spent a couple of hours researching exactly this on Google last week.

The downside to using ChatGPT for research is that it will sometimes make stuff up. In my case, this isn’t a problem since my setting isn’t supposed to be historically accurate. If you do require accuracy, then you might want to double-check ChatGPT’s recommendations.

(Image by Maria Korolov via Midjourney.)

Grammar checkers

I use grammar checkers a lot. My two favorite ones are the built-in AI grammar checker in Google Docs, and Grammarly.

Now, ChatGPT can do something similar.

Let’s write a bad paragraph in the present tense.

so, first, the detective looks arnd. she notices that there’s a bad smell coming from trolls bodie. it has acrid scent. its almost familiar. she thinks back. where has she smelled it befor? she paces away from the body and back again. she looks aorund, searching for inspiration. one of the other trolls rolls his eyes exaspration then pointedly glances at the clock on the temple tower. she doesn’t have much time befor they kick her out of the crim scene.

I want the grammar and punctuation fixed — and let’s say that I want it the past tense, and in the first person. ChatGPT gave me:

I looked around the crime scene, taking in my surroundings. A strong, acrid smell caught my attention and I immediately knew it was coming from the troll’s body. It was familiar, but I couldn’t quite place where I had smelled it before. I paced back and forth, trying to think of any possible leads. One of the other trolls grew frustrated and pointed to the clock on the temple tower, reminding me that I was running out of time. I continued to search the area, determined to solve the case before I was forced to leave.

Yeah, try to get Grammarly to do that.

Also, how exactly did the detective know that the smell was coming from the troll’s body? ChatGPT has some ideas:

  • The detective noticed the smell while she was speaking to other investigators at the crime scene. She looked around to see where the smell was coming from and saw that it was emanating from the troll’s body.
  • The detective noticed the smell while she was examining the other parts of the crime scene. She followed her nose to the source of the smell and saw that it was coming from the troll’s body.
  • The detective asked the other investigators if they had noticed any unusual smells at the crime scene. One of them mentioned the acrid smell coming from the troll’s body, and the detective went over to investigate.
  • The detective noticed the smell while she was interviewing witnesses. One of them mentioned the smell and pointed out that it was coming from the troll’s body.
(Image by Maria Korolov via Midjourney.)

Emotion Thesaurus

The Emotion Thesaurus by Becca Puglisi has been a go-to reference book for me for the past three years.

Say you have a character who’s feeling sad.

You can write something like:

“I have to leave now,” she said sadly.

Or you can replace the word “sadly” with an action or body language that illustrates her sadness. Showing rather than telling, in other words.

I used to keep a cheat sheet of actions that can replace emotions — you can find a copy of it here: Body language master list for writers.

For example, for sadness, I’ve got droopy body, bowed body, wrapping arms around yourself, hesitating movements, bottom lip jutting out, quivering lip, crying, sobbing, shaking, and dragging feet.

Puglisi’s book is infinitely more comprehensive. It also has things like having a puffy face and smudged makeup, wincing, sighing, a slack expression, digging for tissues, feeling like time is slowing down, a need to be alone, loss of appetite, halting one’s speech to gain control, and a whole lot more. Plus, it’s cross-indexed. Sadness may also escalate to depression, for example, or de-escalate to wistfulness.

I love that book.

Now let’s see what ChatGPT can do.

When I asked for some red herrings and higher stakes, it suggested that the pretty elf is at risk if the murder isn’t solved. This will make our detective sad.

Prompt:

in a fantasy world, a hard-boiled detective has to solve the murder of a troll. a pretty elf is helping, but the elf’s own life is at risk if the murder isn’t solved. how can I show that the detective is sad without saying the word “sad”? I want to have some actions in the story that can demonstrate her sadness. but the detective doesn’t like to show her emotions. so the actions have to be understated.

ChatGPT:

There are a few ways you can show that the detective is sad without saying the word “sad” or having her express her emotions overtly. Here are a few ideas:

    1. You can describe her body language and facial expressions: She might look downcast or dejected, or her posture might be slumped.
    2. You can show her withdrawing from social situations: She might not want to be around other people or might avoid socializing.
    3. You can describe her behavior as being more subdued or quiet: She might not be as energetic or talkative as she usually is.
    4. You can show her engaging in self-care activities: She might spend more time alone, perhaps taking long walks or indulging in comfort foods.
    5. You can describe her having trouble concentrating or sleeping: She might have difficulty focusing on work or have insomnia.

Remember, there are many ways to show emotion without saying it outright, so try to be creative and think about how the detective might behave if she were feeling sad.

I think I want more body language suggestions.

Prompt:

can you expand on option one?

ChatGPT:

  • “She looked down at the ground, her shoulders slumped as if the weight of the world were on them.”
  • “Her face was drawn and tired, her eyes rimmed with red as if she had been crying.”
  • “She kept her arms crossed over her chest, hugging herself as if trying to find comfort in her own embrace.”
  • “Her lips were pressed into a thin line, her brow furrowed as if she were deep in thought or struggling with some inner turmoil.”
  • “She didn’t make eye contact with anyone, her gaze fixed on the floor as she walked through the room.”
  • “She sat at her desk, her arms resting limply at her sides and her head tilted downward as if in defeat.”
  • “Her eyes were glassy and distant, as if she were lost in some private grief.”
  • “She fiddled with her wedding ring, twisting it round and round her finger as if seeking comfort in the familiar motion.”
  • “She walked slowly and deliberately, her steps heavy and dragging as if she were carrying a burden.”
  • “Her voice was soft and lacked its usual conviction, as if she had lost faith in her own abilities.”
  • “Her hands trembled slightly as she reached for her coffee cup, her normally steady grip now unsteady.”
  • “Her eyes were red-rimmed and puffy, as if she had been crying or was struggling to hold back tears.”
  • “She avoided making eye contact with anyone, her gaze fixed on the floor as if she were too overwhelmed to look at anyone.”
  • “She hugged herself tightly, her arms wrapped around her body as if she were trying to hold herself together.”
  • “Her smile was forced and didn’t reach her eyes, which seemed clouded with sadness.”

Plenty of options here that I can tweak and use for the story.

(Image by Maria Korolov via Midjourney.)

Online translators

If I want to have your characters say things in foreign languages, I typically use Google Translate. If it’s an usual language, I’ll use a specialized tool, like the English to Shakespearean translator if I want a medieval flavor for a piece of dialog. There are also some not-too-great Klingon translators online.

ChatGPT can handle pretty much any translation — including Klingon. And the results sound natural and authentic. Plus, it will give you explanations. For example, here’s what it gave me for how to say “Where is the bathroom?” in Klingon.

ChatGPT:

In Klingon, the phrase “Where is the bathroom?” can be translated as “nuqDaq ‘oH puchpa’ ‘e’ yIlaD?” (pronounced “nook-dak oh pooch-pah ay yee-lad”).

Here is a breakdown of the individual words in this phrase:

    • nuqDaq: where
    • ‘oH: is
    • puchpa’: bathroom
    • ‘e’: the (direct object marker)
    • yIlaD: located

Keep in mind that Klingon is a constructed language and is not spoken natively by any group of people. It was created for use in the Star Trek franchise and is used primarily in that context. While it is possible to learn and speak Klingon as a hobby or as part of a role-playing activity, it is not a widely spoken language and you may not find many people who are able to understand or speak it fluently.

I double-checked online, and it’s correct. It did an equally good job with the Russian translations I requested, complete with notes about the cultural context of the phrases and pronunciation tips.

(Image by Maria Korolov created with Midjourney.)

Beta readers

Beta readers are hard to find, especially when you’re first starting out.

I have a writing mastermind group that meets once a week. I’m lucky. Not everyone has that. Or maybe they write so much that their mastermind group can’t keep up.

Can ChatGPT help? Yes, it can!

I gave it the prompt: “Tell me what you think of the following text. Do you see any plot holes? Are the characters sufficiently motivated? Are there places where I should show instead of tell? Does my writing drag anywhere? Are there any parts that are confusing?”

Then I pasted in a snippet of a story I’m currently working on.

ChatGPT:

Overall, the text appears to be describing a scene where Eggatha is getting ready to drive to the train station to meet her granddaughter and her two children. It seems that she is being assisted by George Pen, who is helping her prepare to drive a pickup truck. The text mentions that Eggatha used to drive a standard transmission car back in the “old days,” but she hasn’t driven much since moving to Hereabout Grid. There is also mention of a butler named Sam Wilson who could potentially drive her instead. The text describes Eggatha’s thoughts as she prepares to drive and there is some dialogue between her and George.

There are no plot holes that I can see in the text. However, it is not entirely clear what the motivation is for Eggatha to be driving to the train station. It is also not clear why she is driving a pickup truck rather than a car. It might be helpful to provide more context or motivation for the characters’ actions.

There are a few places where the text could be improved by showing rather than telling. For example, instead of saying “Eggatha wasn’t Eggatha Crispie back then, of course,” it might be more effective to describe the circumstances under which she became Eggatha Crispie. Similarly, it might be more interesting to describe Eggatha’s experience driving a standard transmission car back in the “old days,” rather than simply stating that she used to do so.

The writing does not appear to drag anywhere in the text. However, there are a few parts that might be confusing for readers. For example, it is not clear what “Hereabout Grid” refers to or why Eggatha moved there. Additionally, the reference to a murder in the village and the possibility of a serial killer might be confusing without more context. It might be helpful to provide more information about the setting and the events taking place in the story to help readers better understand what is happening.

It was totally right. The first paragraph is a perfect summary of the scene.

ChatGPT was also right in suggesting that I didn’t make Eggatha’s motivation clear, and why she’s not driving her car instead of the pickup. And it was right about the other points it raised, as well.

Frankly, I’m a little creeped out.

I put in the first chapter of Pride and Prejudice. Here is what it had to say:

There do not appear to be any plot holes or confusing parts in the text. The characters are motivated by their desire for social status and financial security, which is clear from the conversation between Mr. and Mrs. Bennet. The writing does not drag and the plot is easy to follow. The characters’ motivations and desires are shown through their dialogue and actions, rather than being told directly to the reader. Overall, the text is well-written and engaging.

Of course, I was asking it to comment on a rough draft of my writing. Of course, it wasn’t polished yet. Let’s ask it what it thinks of the first chapter of my book Krim Times, which went through several rounds of editing.

Here is what it had to say

It’s difficult for me to provide a detailed analysis of the chapter without more context or information about the larger story. From what is presented in this chapter alone, I don’t see any obvious plot holes or areas of confusion. The character of Marshal Henderson Trask is introduced as being involved in the investigation of the sniper, which provides sufficient motivation for his actions in this chapter. The descriptions of the virtual world and its inhabitants also provide a sense of setting and context for the events that are unfolding.

In terms of the writing itself, there are a few places where you could consider showing rather than telling in order to bring the story to life for the reader. For example, instead of simply stating that “Murder and robbery were not, as such, against the grid’s terms of service,” you could consider describing how these crimes are actually carried out or how they are tolerated within the virtual world. This would help to bring the world and its inhabitants to life for the reader and make the story more immersive.

I didn’t notice any specific areas where the writing dragged or felt slow. However, as a general suggestion, you might consider varying the length and structure of your sentences in order to keep the reader engaged and maintain a good pace. This can be achieved through the use of shorter, punchier sentences to create a sense of momentum, as well as longer, more descriptive sentences to provide detail and depth.

Overall, this chapter seems to be well-written and effectively establishes the setting and introduces some of the main characters and conflicts that may be developed further in the story.

Not too bad! Thanks, ChatGPT!

Here’s a tip for using it as a beta reader:

Say, “Tell me what you think of this chapter:” then paste in the text of the chapter and press return.

Then ask follow-up questions:

  • Are there any plot holes?
  • Are there any places that don’t make sense?
  • Are the character motivations believable?
  • Are there any places where the writing drags?
  • Are there any places I should show instead of tell?
  • Are there any places where the grammar or style needs improving?
  • Are there any other ways in which the story can be improved?

I just asked ChatGPT to look at a few chapters that friends wrote, and the advice wasn’t particularly comprehensive and it did miss some glaring issues. But it did an excellent job of summarizing their text and identifying the major themes.

Do you have any other ideas for questions to ask ChatGPT to help it be a better beta reader? Let us know in the comments!

(Image by Maria Korolov via Midjourney.)

Therapy

My therapist has changed my life. Ironically, for someone who writes for a living, I’m bad at expressing my own emotions and opening up to people. I’m also someone who sets too-high standards for herself, and when my work doesn’t measure up, I just stop trying.

My therapist helped me with these issues, and with other personal stuff. I highly recommend having someone to talk to about the deep things going on in your life. And if you can’t open up to family or friends, find someone else you can talk to.

I wrote nearly a dozen novels and novellas with the help of my therapist — now I just need to talk to her about getting them edited and published. What is holding me back now? Who knows! But she can help me find out!

Meanwhile, let’s see if ChatGPT can help me get more stuff done tomorrow.

Prompt:

I’m feeling stuck on my latest book. I’m having trouble motivating myself to write.

ChatGPT:

It can be challenging to stay motivated and focused when working on a long-term project like writing a book. Here are a few strategies you might try to help you stay on track:

    1. Set specific, achievable goals for each writing session. Rather than trying to write the entire book at once, try breaking it down into smaller, more manageable chunks. This can help you feel more motivated and make it easier to stay focused.
    2. Find a dedicated workspace. Having a specific place to work can help you get into the right mindset for writing. Make sure your workspace is comfortable and free of distractions, so you can focus on your work.
    3. Set aside regular writing time. Try to schedule the same time each day to work on your book. This can help you develop a routine and make it easier to get into the habit of writing every day.
    4. Take breaks. It’s important to take breaks to rest and recharge. Taking regular breaks can help you stay focused and motivated over the long term.
    5. Seek support. Consider joining a writing group or finding a writing partner to help keep you accountable and motivated. It can also be helpful to talk to friends or family members about your goals and ask for their support.
    6. Find what works for you. Everyone is different, so it’s important to figure out what works best for you. Experiment with different strategies and approaches until you find what helps you stay motivated and focused on your writing.

Me:

I have a support group. and I sit down to write at 8 a.m. every morning in the same place at the same time. How can I tell if there’s an emotional reason holding me back?

ChatGPT:

If you’re finding it difficult to stay motivated and focused despite having a support group and a dedicated writing space and time, it’s possible that there may be an emotional issue that is holding you back. Some common emotional barriers to writing include:

    1. Fear of failure: You may be afraid that your writing will not be good enough, or that you’ll never finish your book.
    2. Procrastination: You may be avoiding writing because you’re not sure where to start, or because you’re afraid of writing something that isn’t perfect.
    3. Lack of confidence: You may doubt your ability to write a book, or feel like you’re not qualified to write on the topic you’ve chosen.
    4. Perfectionism: You may be putting too much pressure on yourself to write the perfect book, and as a result, you may be paralyzed by the fear of making a mistake.

If you think that there may be an emotional issue holding you back, it might be helpful to try some techniques to help you identify and address those feelings. For example, you could try journaling about your feelings and concerns, or talking to a therapist or coach who can help you work through any emotional blocks you may be experiencing.

Me:

Maybe it’s fear of failure. How can I tell?

ChatGPT:

There are a few signs that fear of failure may be holding you back:

    1. You’re afraid to start writing because you’re worried that your work won’t be good enough.
    2. You have a hard time completing your writing because you’re afraid that it won’t meet your standards or the standards of others.
    3. You’re afraid to share your work with others because you’re worried about their criticism.
    4. You feel overwhelmed by the thought of writing a whole book and are afraid that you won’t be able to finish it.

If you’re experiencing any of these feelings, it may be helpful to try some strategies to help you overcome your fear of failure. For example, you could try setting small, achievable goals for each writing session and celebrating your progress as you meet those goals. You could also try reminding yourself that it’s okay to make mistakes and that writing is a process that involves revising and editing. It might also be helpful to talk to a therapist or coach who can help you develop strategies for dealing with your fear of failure.

Me:

That sounds about right. Can you suggest some ways to celebrate my progress after achieving small goals?

And ChatGPT responded with some very useful advice.

Next time, I might want to discuss my relationship with my mother.

How are you using ChatGPT? What features do you want to see in the commercial version, once it comes out of beta? What are your favorite prompts? Let us now in the comments below!

 

Edited by Melody Friedenthal

MetaStellar editor and publisher Maria Korolov is a science fiction novelist, writing stories set in a future virtual world. And, during the day, she is an award-winning freelance technology journalist who covers artificial intelligence, cybersecurity and enterprise virtual reality. See her Amazon author page here and follow her on Twitter, Facebook, or LinkedIn. Email her at [email protected]. She is also the editor and publisher of Hypergrid Business, one of the top global sites covering virtual reality.

6 thoughts on “11 writing aids that will be replaced by ChatGPT”

  1. We are rapidly approaching a future where we are no longer creators but curators; picking from a selection of pieces offered up by our AI generators because it’s far easier than making anything ourselves.
    And I for one am terrified.

    1. Dave — I totally agree. We are approaching the point where we cannot predict what the future will be — we just know that it will be radically different.

      I have a feeling it will be the same kind of sea change as the industrial revolution. Before, the vast majority of us were subsistence farmers. Feeding our families was our top priority. Food scarcity was an issue everywhere on the planet. One bad season could wipe out a harvest and result in famine. If we needed anything, with few exceptions, we needed to make it ourselves — clothing, furniture, houses, entertainment. The only people who didn’t live this way were the nobility. They took stuff away from the people who produced it, generally with violence — either as payment to protect people from others’ violence, or by directly using violence themselves. And when they nobility took stuff, they either used it up themselves, or stored it in their castles. Either way, the stuff they took became inaccessible to everyone else.

      Industrialization resolved the problem of stuff. We could now make more stuff than we needed. With the exception of places undergoing political upheaval, or people with unmet physical or mental needs (also a political issue, not a practical one) there’s no shortage of food or other stuff in the world today. There’s a shortage of political common sense, but that’s a different issue.

      Today’s nobility — i.e., rich people — no longer take stuff away from everyone else and hoard it. They make their money by GIVING stuff to people in return for little piece of paper. And when they collect the paper they don’t extract it and hoard it and keep it from being used by everyone else — they put it in the bank or stock market. That means that the money they extract from people goes right back into circulation — banks loan it out to people for buying houses and cars, to get educations, and to start businesses.

      Rich people are still a pain in the butt, but they’re no longer the drain on civilization they once were. The gap between rich and poor is higher than ever — but only on paper. In practical terms, the gap between rich and poor has never been smaller. A middle-class person can save up and travel anywhere in the world. They get roughly the same medical care as rich people do — rich people can afford to pay for experimental treatments, but, as far as I’m concerned, I’m happy to let them be the guinea pigs and I’ll wait for the FDA to approve the stuff and for my insurance to cover it. Middle-class people watch the same entertainment and eat pretty much the same food and wear the same clothes as rich people do. Middle-class people have nurses and doctors and teachers and they can get food delivery and housekeeping services — stuff that only rich people had before the industrial revolution.

      For people who love farming, it is harder to do it for a living. You have to market yourself well to compete against agribusiness. Each of your cows has to have its own Facebook page. But its doable – and people farm out of love, not out of necessity.

      We’re now entering a world where AI can write its own code and improve itself — and where we are editing our own DNA to eliminate diseases. We have a space industry. We’ve cracked nuclear fusion (though it will take a couple of decades to commercialize it). The future isn’t guaranteed. We can still have environmental catastrophes, economic meltdowns, pandemics, or an asteroid can crash into the planet and kill all life. But, assuming that none of that happens before we can take measures, the future will look very different from the present.

      And we have no way of knowing how. What will happen when it’s not just stuff that’s in plentiful supply, but work as well? If AIs and robots can do all our work for us, what does that mean for how we will live?

      My guess is that the following will happen:

      * We will get universal basic income. Rich countries first, then everyone else soon after. The gap between the developing and developed world has never been smaller, and AI will accelerate the pace of change everywhere. Some jurisdictions are already experimenting with UBI. Not to mention the stimulus checks we got during the pandemic.

      * We will no longer have to do menial work in order to feed and clothe ourselves. We will do work because we want to. We will still have artists and writers, but they will no longer have to depend on selling their work in order to survive.

      * As social animals, we will continue to value human contact. Jobs that involve the personal touch, personal attention, or personal connections will still be plentiful. These include jobs in the field of medicine, law, business consulting, coaching, financial advice, personal care such as massage and food service, entertainment, sports, and many more.

      * As humans, we will still have goals and motivations. Jobs that involve achieving those goals will still be plentiful. For example, we want to clean up the environment. There will be plenty of jobs to help do that, even if some of those jobs will involve managing AIs doing the actual hard work.

      * As lifespans expand and we stay healthier longer, retirement will be replaced by sabbaticals. We will save in order to be able to take a few years off to find ourselves — to do art, to commune with nature, to go back to school for new degrees, to volunteer, to spend time with family, or just sit around and catch up on TV and video games.

      * As birth rates drop around the planet, governments will do more to encourage having children. Finally — finally! — providing support for parents in the form of childcare, time off, and other resources. This is already starting to happen in countries with the lowest birth rates. As more countries move up the economic ladder, this trend will continue.

      I personally think this new world will be infinitely better than the current one, just as post-industrialization life was infinitely better than pre-industrialization. There will be lots of downsides, especially during the initial transition period. Lots of people and industries will be hurt, and it will be a good idea to plan ahead in order to minimize the damage.

      But the cat is out of the bag. The ball is rolling. We’ve driven onto the onramp to the post-industrial cyberhighway and there’s no turning around and getting off.

      Our brains aren’t wired to think about non-linear change. I myself am still saving for retirement, still worrying about my kids’ careers, still doing everything I should to prepare for a world that’s exactly like our current one, just a little bit different.

      We can’t plan for a world that’s radically different — we’re not wired that way. It doesn’t make any sense for us to do that. We automatically dismiss any prediction out of hand that doesn’t match current reality.

      But the AI revolution will be as big as the industrial one, and that means that the changes we will see will be as big as those, if not bigger. Our entire economic system will completely change. The things we worry about today will become irrelevant. Things currently impossible will not only become possible, but probable.

      I’m lucky that I get to write about all this in my day job, and interview people who are making it happen. I’m getting a front-seat view. But all the information is accessible to anyone — there are millions of YouTube videos, podcasts, and publications chronicling the changes. None of them should come as a surprise. Of course, given the way our brains are wired, it will all be a surprise — and the changes will be frightening. There will be pain and resistance.

      I think that’s why some people are getting so angry about AI art and AI text. Having AI create a picture doesn’t stop an artist from also creating a picture — just like putting cameras in people’s hands didn’t stop other people from painting portraits. We can have both. And we now have a new profession, too — photography. We will have a profession of AI artist. Creating AI art can be annoying and time consuming. Sometimes, you’ll just want to have someone else do it for you — just like you can take your own pictures at your wedding, or you can hire a photographer.

      But even if you take your own pictures, they will still be infinitely better than if all the pictures had to be drawn by hand and people had to pose for hours for each one.

      AI art and AI text will raise the lowest standard. We will have a new minimum level of quality, the level that’s enabled by AI. And humans will look for opportunities to do better, or to do different, then what AI offers. I think that’s going to be the sweet spot. Use AI to do the routine work, then apply your own creativity and personality and values to add a layer on top of that.

  2. Maria, this is such a great article about ChatGPT, including this comment of yours about how you see our future with AI. I’m adding a little piece of information about your very last comment about Jasper. Jasper and ChatGPT are the same thing. There’s no competition (as far as technology is concerned). This is because Jasper is just a front-end using GPT-3 via API requests. Jasper is really expensive. I personnaly use WordHero which is also based on GPT-3 but got a one-time purchase deal via Appsumo. The vast majority of AI writing tools are actually all based on the same backend company: OpenAI. GPT is a technology released by OpenAI and its current last version is version 3 and is only accessible via API. ChatGPT is the version 3.5 adding the chat usage that is being more user-friendly. I don’t know if the commercial release will only be accessible via API or if OpenAI will release a commercial app either desktop or mobile even non developers could use.

  3. Maria, this is such a great article about ChatGPT, including this comment of yours about how you see our future with AI. I’m adding a little piece of information about your very last comment about Jasper. Jasper and ChatGPT are the same thing. There’s no competition (as far as technology is concerned). This is because Jasper is just a front-end using GPT-3 via API requests. Jasper is really expensive. I personnaly use WordHero which is also based on GPT-3 but got a one-time purchase deal via Appsumo. The vast majority of AI writing tools are actually all based on the same backend company: OpenAI. GPT is a technology released by OpenAI and its current last version is version 3 and is only accessible via API. ChatGPT is the version 3.5 adding the chat usage that is being more user-friendly. I don’t know if the commercial release will only be accessible via API or if OpenAI will release a commercial app either desktop or mobile even non developers could use.

Leave a Reply to Maria Korolov Cancel Reply

Your email address will not be published. Required fields are marked *