Category: AutoCrit

AI Editor Issues

Posted in Artificial Intelligence, AutoCrit, ChatGPT, Claude, critique, Generative AI, Short Story, Software, WritingTagged , , , , , , , , , , Leave a comment

I employ AI editors for copyediting and alpha-reading. They are useful but have limitations.

Some of my writing is ordinary – Acts I, II, and III; Beginning, Middle, and End. This is AI’s sweet spot: assess a piece and compare it to a million similar pieces, sharing plot structures, story and character arcs, heroes’ journeys, and saving cats.

Other stories are experimental. They don’t follow the Western tradition of tidy storylines and neat little bows, evey aspect strongly telegraphed, so as not to lose any readers along the journey.

Mary approaches a doorway. Mary opens the door. She walks through the doorway — the doorway she had approached.

Obviously, this is silly and exaggerated, but the point remains. AI presumes that readers need to be spoonfed, especially American audiences. (No offence.)

But life doesn’t work like this. We often witness events where we have no idea what happens after we experience them. We pass strangers on the street, not knowing anything about their past or future. We overhear something interesting, never to get a resolution. We get passed by for a promotion but never know the reason why.

In science, there are lots of dead ends. Do we want to know the answers? Yes. Is one likely? Maybe; maybe not. Will we make up answers just to satisfy our need for closure? It happens all the time.

In writing, we seem to not accept these loose ends. How many times have you read a review or critique where the complaint is, “What happened to this character?” or “Why didn’t Harry Potter use his invisibility cloak more than once despite it being an obvious solution to many prior and future challenges he faced?”

Sure. I agree that it feels like a plot hole, but the author doesn’t have to tell you that Harry lost it in a poker match, it got lost in the wash, or Ron snatched it.

I’m finishing a story, and various AIs provide similar commentary. Even more humorous are the times it can’t follow a thread, but when a human reviewer reads it, they have no difficulty. In the end, there may be unanswered questions. Some of these leave the universe open for further exploration, but not all questions have answers. AI has difficulty grasping this perspective.

AutoCrit Challenges

Posted in Artificial Intelligence, AutoCrit, ChatGPT, Claude, critique, Generative AI, Short Story, Software, WritingTagged , , , , , , , , , , , , , , , , , , , , , , , , , , Leave a comment

I don’t hide the fact that I rely on AI for early editorial feedback. Once a story is complete, I break out AutoCrit. This programme works well for typical stories that follow standard practices with common tropes. It gets quite confused when I feed it intentionally awkward stories, not the least of which is to advise me to eliminate the awkwardness.

This is a challenge with AI more generally. In this particular story, I leave a lot of loose ends and misdirects, as it’s a commentary on the conspiracy-driven culture we inhabit. The advice, is along the lines of, “You forget to close this lopp. What happened to so and so.”

But this is life. We don’t always know the full story. We drive past an multi-car accident where cares are overturned and in flames, but we never find out what happens – even if we scour the newspapers and internet. Who was that? What happened? What caused it?

We often never find out. In most books and movies, we find out everythung, and it all comes packaged with a nice bow. This is what AI expects. It’s the diet it’s been fed.

Some stories subvert these notions here and there, but by and large, this is not typical American fare. Readers and viewers need to be spoonfed without inconsistencies.

Speaking of inconsistencies addressing one scene, AutoCrit said that a character should act impulsively in one situation and reserved moments later. This was flagged as an iinconsistent character.

In the scene, a woman stops her car immediately to help an injured man on the roadside, but as she gets out of her car an approaches her, she shows caution.

This was a red flag. Why would she have always been rash or always been cautious?

My response, because that how real people act. She acts on instinct but quickly considers that she’s a vulnerable woman alone with a man miles from anywhere.

I don’t suspect a human reader would find this surprising. This is the intelligence absent from Artificial Intelligence — cultural intelligence, a cousin of EQ, emotional quotient.

I know how I want the character to act. I do want AutoCrit to inform me that character A is wielding a pistol but then stabs another character, or that character B is a teetotaler and is getting drunk or that character C has a shellfish allergy but is downing lobsters like they’re going out of style. And I certainly what to be shown continuity errors.

The biggest challenge I have with AutoCrit that is less promonent with other AIs is that I can preface my content with a note explaining my intent. I can even do this after the fact.

If I feed ChatGPT, Claude, or DeepSeek a story of segment to critique without a preface, the responses may be similar to AutoCrit, but when I follow up with some meta, the response may be, “Now it makes sense, but why is John wearing lipstick?” Perhaps he’s metrosexual or non-traditional. Perhaps it’s an oversight.

I dont meán to demean AutoCrit. I’m just advising that if you are writing stories not compliant with 80 per cent of published works, take the advice with a grain of salt, or reserve AutoCrit for more standard fare.

Is it AI?

Posted in Artificial Intelligence, AutoCrit, ChatGPT, Claude, critique, Generative AI, Short Story, Software, WritingTagged , , , , , , , , , , , Leave a comment

I favour originality even at the expense of popularity or sales. I spent last week writing short stories and poems. I use AI for research, whereas in the “old days”, I’d have used a library. I research character traits and arcs, story forms, and whether a theme has been explored.

I employ AI in the editorial process, and even in “post-production”. I even use AI for some art concepts and components.

One thing I hadn’t tried until now is an AI service that purports to determine if a submission is AI. I tried several packages that offered a free trial. They seem to operate on a scale between human and AI authorship.

I first submitted a piece I was currently working on—a 6th-odd revision of a 5,000-word story in the form of a fairy tale. Unfortunately, trials were limited from a sentence to a few paragraphs—up to 5,000 characters.

This first submission was rated 100% AI—evidently, not a hint of humanity. This was disconcerting. I decided to dredge out a non-fiction book I shelved in 2020. Certainly before access to AI tools. This was rated 85% AI and 15% human. But it gets better—or worse, I suppose, depending on your perspective.

The book is on the immorality of private property from a philosophical vantage. The passages claimed to be AI were one-hundred per cent mine. What about the ones flagged as human, you might be asking? Those were a quote by fellow human John Locke from his Second Treatise of Government.

In Defence of Property 

“God, who hath given the world to men in common, hath also given them reason to make use of it to the best advantage of life, and convenience. The earth, and all that is therein, is given to men for the support and comfort of their being. And though all the fruits it naturally produces, and beasts it feeds, belong to mankind in common, as they are produced by the spontaneous hand of Nature; and no body has originally a private dominion, exclusive of the rest of mankind, in any of them, as they are thus in their natural state: Yet being given for the use of life, there must of necessity be a means to appropriate them some way or other before they can be of any use, or at all beneficial, to any particular men.”

ᅳ John Locke, Second Treatise of Government 

Returning to the AI side, what sentences were flagged as the “Top Sentences driving AI probability”? I’m glad you asked.

  • The Catholic Church also played a significant role in shaping private property rights in the Middle Ages.
  • In ancient China, the concept of private property was more limited, as land was owned by the state and was leased to individuals for use.
  • However, there is evidence to suggest that private property ownership has existed in some form in many ancient civilisations.
  • Although it’s difficult to trace the precise history of private property ownership before ancient Greece, the concept of private property has evolved over time and has varied widely among different societies.
  • It regulated the transfer of property and established rules for inheritance.

So these ordinary sentences written 5 or more years ago are flagged as AI.

The US Constitution

On a site I found to understand what parameters AI considers, I found this example—the Constitution of the United States of America was flagged as having AI content. I knew those geezers were ahead of their time, but I didn’t realise how far. This is even more amazing when one considers that electricity hadn’t even yet been invented.

But Why?

AI looks for statistically probable patterns. This translates into any content written with proper grammar and diverse word choice. In practice—the habits of a decent writer.

I’m not going to belabour this issue, but I want to raise a big red flag.

To complicate matters more, they have AI applications that promise to un-AI your AI. So there’s that.

AutoCrit Innards

Posted in Artificial Intelligence, AutoCrit, Ridley Park, Short Story, Software, Status Update, WritingTagged , , , , , Leave a comment

Writing is hard. Short stories are worse. I started Mind Without a Mirror a few days ago as a short story project. After a dozen major revisions, I got to a place to run it through AutoCrit. I’ve been using AutoCrit for a couple months, and it’s been useful as an editor before I connect with a human editor or Beta reader.

Today, I think it split its guts. I clicked on the Character tab. This is where it assesses your character traits, strengths, weaknesses, and some other aspects. As you may notice for the first character, Ada, it returns a terse response. This is usual. The second character Echo went off the chain.

Major characters including Ada and Echo provide contrasting perspectives aiding in highlighting different facets of conflict surrounding Sol’s disappearance:

1. Ada – Her impulsiveness offers tangible counterpoints but sometimes lacks depth behind motivations driving rash decisions; deeper backstory integration can enrich relational dynamics while avoiding plot holes associated with seemingly arbitrary choices leading toward unnecessary risk-taking scenarios without sufficient narrative justification.

2. Echo – As a voice urging caution yet pushing boundaries intellectually rather than physically contrasts effectively against both Ada’s impulsiveness and initially hesitant nature exhibited by Nova; further scenes emphasizing logical deductions alongside emotional intelligence contributions can elevate effectiveness within group dynamics exploring unknowns collectively ensuring smoother narrative cohesion devoid apparent gaps particularly during critical junctures necessitating unanimous decision-making processes amongst protagonists’ circle thereby mitigating potential dissonance arising from conflicting individual agendas undermining collective objectives pursuit efficiency notably during climax build-up phases preceding resolution stages inherently reliant upon concerted efforts fruition realizing overarching goals set forth early onset storyline unfolding sequence events trajectory mapping course eventualities encountered en route denouement culmination point reached conclusionary chapter segments encapsulating thematic essence distilled core message intended conveyed audience reception interpretation thereof facilitated 

I shared a screenshot so you can see the random word dump. Perhaps it’s speaking in tongues. Toward the bottom of the laundry list, I see a lot of professional titles below some superlatives.

I don’t know. AI is strange. I wasn’t planning to post anything today, but I just had to share.