Tag Archive | AI-Assisted Writing

When AI Repeats Itself: Why Good Journalism Needs More Than Accurate Information

By Cheval John | Vallano Media
This article was created with the assistance of artificial intelligence.
The final version was reviewed, edited and fact-checked by the author


Graphics created with AI-assisted design tools.
Visual concepts, editorial direction, composition, and final creative decisions were developed by Vallano Media.

One of the lessons I have learned from working with artificial intelligence is that an article can contain accurate information and still need substantial editing.

When I began developing my AI-assisted editorial workflows at Vallano Media, I placed considerable attention on verifying information.

I wanted to make sure the claims in my articles were supported by evidence and that the finished work met my standards as a journalist and publisher.

But as I continued experimenting, I encountered another problem.

Sometimes, an article would repeat the same argument several times, even though the wording was different.

I would read through a draft and realize that a paragraph was explaining something the article had already established.

Another section might introduce the same idea using slightly different language without providing additional evidence or analysis.

The information was not necessarily incorrect.

The problem was that the article was repeating ideas instead of developing them.

When I noticed this happening, I would instruct the AI assistant to rewrite the material and eliminate the unnecessary repetition.

That experience reminded me of something important about journalism.

Checking the facts is essential.

But producing a good article also requires knowing how to organize those facts, develop an argument, and communicate with readers without wasting their time.

When an Article Says the Same Thing More Than Once

Repetition has always been an editorial concern.

Human writers can repeat themselves just as AI-assisted drafts can.

The difference in my experience was that I was learning to recognize and address the problem while working with a technology capable of generating and revising large amounts of text quickly.

Consider a hypothetical article about how artificial intelligence can help small businesses improve their operations.

One paragraph explains that AI can automate routine administrative tasks.

Another explains that business owners can save time by using AI for repetitive work.

A third explains that automation allows employees to concentrate on other responsibilities.

These ideas are related, but they are not necessarily identical.

If the second paragraph introduces evidence about time savings and the third examines how employees use that additional time, each paragraph contributes something meaningful.

But if all three paragraphs simply repeat that AI saves time, the article has a problem.

It is becoming longer without necessarily becoming more informative.

This distinction matters because removing repetition should not mean removing useful information.

The objective is to develop the article, not simply reduce its length.

Repetition Is Not Always a Mistake

Not every repeated idea needs to be eliminated.

Sometimes a journalist returns to a central argument because the reader needs to understand its importance.

An introduction might establish a problem.

The body of the article might examine that problem through evidence and examples.

The conclusion might return to the original argument after the reader has gained a deeper understanding.

That is not necessarily unnecessary repetition.

It can be part of a deliberate structure.

The question is whether the repeated idea contributes something new.

Does it introduce additional evidence?

Does it provide context that was missing earlier?

Does it explain a consequence or offer a different perspective?

Or is the article simply repeating something the reader already understands?

This is one of the distinctions I have to make when reviewing AI-assisted writing.

I do not want an assistant removing every repeated reference to a central theme.

That could weaken the article’s coherence.

Instead, I want the writing to develop the subject while maintaining a clear connection to the original argument.

Recognizing the Problem as an Editor

When I encountered repetition in my AI-assisted articles, I did not need to conclude that the entire draft was unusable.

I needed to identify what was wrong with the writing.

Sometimes the problem was a paragraph that restated an earlier point.

Other times, it involved sections that covered similar ground without advancing the discussion.

My response was to direct the AI assistant to revise the material.

That might seem like a small decision, but it illustrates an important part of my editorial process.

The assistant could produce another version of the article.

I could also ask it to identify repetitive passages for further review.

But I was the one evaluating whether the article communicated what I intended.

That responsibility involved more than identifying repeated words.

Two paragraphs can use completely different language while expressing essentially the same idea.

A draft may appear polished because the sentences are grammatically correct and the transitions sound professional.

Yet the article can still lack progression.

Recognizing that problem requires reading beyond the individual sentences and examining the overall argument.

I have learned to ask a straightforward question:

“Is this paragraph giving the reader something new, or is it simply saying something the reader already knows?”

That question helps me determine whether a passage deserves to remain in the article.

Revising Without Losing the Argument

Once repetition has been identified, the next challenge is correcting it.

Artificial intelligence can help by combining similar paragraphs, reorganizing sections, or suggesting more concise language.

Those capabilities can make certain editorial tasks faster, although the resulting revisions still require careful review.

The danger is that a revision might remove something important.

A paragraph that appears repetitive could contain a qualification that changes the meaning of a claim.

Another might include evidence that supports the argument.

If those details disappear during rewriting, the article could become shorter but less accurate or less informative.

This is why I cannot judge a revision solely by its length.

I have to consider whether the revised version preserves the intended meaning, necessary context, and supporting information.

I also have to make sure the article still sounds like something I would publish.

There is a difference between improving the writing and allowing the technology to change the direction of the article.

The first supports my editorial work.

The second could undermine it.

Human editors are not immune to these problems.

I can overlook repetition or make a revision that weakens a passage.

Recognizing that possibility is another reason to review changes carefully rather than assume that either human or AI judgment is automatically correct.

The objective is not perfection through a single revision.

It is a process that provides opportunities to recognize and correct weaknesses.

Accuracy Is Not the Same as Editorial Quality

This experience also helped me understand why an AI-assisted newsroom needs different kinds of editorial review.

An article can be factually accurate while containing unnecessary repetition, confusing transitions, or poorly organized arguments.

It can also be well organized and enjoyable to read while containing unsupported claims.

Both situations require attention, but they present different editorial problems.

At Vallano Media, Truth Review examines factual claims and their supporting evidence.

Red Team Review challenges assumptions, reasoning, and weaknesses that might not have been addressed during earlier stages.

Other editorial reviews focus on how the article communicates its message, including structure, clarity, and consistency with my intended voice.

These responsibilities can overlap, and an issue identified during one review may require further examination during another.

Each stage nevertheless serves a different editorial purpose.

For example, a Truth Review might establish that a statement is supported by reliable evidence.

That finding does not automatically mean the statement needs to appear in multiple sections.

Likewise, an editorial revision might eliminate unnecessary repetition without establishing whether the remaining factual claims are correct.

This is why I do not treat one successful review as proof that the entire article is ready for publication.

The reviews work together, but the final editorial decision remains my responsibility.

Respecting the Reader

One reason I take repetition seriously is that readers give their time and attention to the articles I publish.

They should receive information that helps them understand the subject.

That does not mean every article must be short.

Some subjects require extensive explanation, historical context, supporting evidence, and careful analysis.

A longer article can be appropriate when its length serves the story.

But length alone does not establish depth.

An article that repeats the same point five times is not necessarily more informative than one that explains the point clearly and then develops the discussion.

For me, this is an important distinction as an independent journalist and publisher.

I want Vallano Media’s articles to be accurate, but I also want them to communicate effectively.

That means considering what the reader already knows from earlier paragraphs and what information should come next.

Sometimes the best editorial decision is to expand a section because the reader needs more context.

Other times, the best decision is to remove a paragraph because the explanation has already accomplished its purpose.

The goal is to respect the reader without sacrificing the substance of the article.

Beyond the First Draft

Working with artificial intelligence has changed how I approach the development of an article.

I no longer look at an AI-generated draft simply as something to accept or reject.

I look at it as material that can be examined, challenged, and improved.

When I encounter repetition, I can identify the problem and ask the assistant to revise the writing.

When a factual claim needs verification, I can return to the supporting evidence.

When an argument lacks clarity, I can reconsider how the article is organized.

These are different editorial decisions, but they share a common principle.

The technology assists with the work.

It does not eliminate the need to decide what the work should accomplish.

That distinction is especially important as I continue developing my AI-assisted newsroom.

I am not trying to create a system that produces articles without human involvement.

I am building a process in which artificial intelligence supports research, writing, verification, revision, and publication while I retain responsibility for the final product.

My experience with repetitive writing may appear to be a relatively small example of that responsibility.

But it demonstrates why editorial judgment matters even when the facts themselves are not in dispute.

I recognized that the writing was repeating itself.

I directed the assistant to revise it.

And the responsibility for deciding whether the revised article met my standards remained with me.

Producing another draft is one task.

Deciding whether that draft communicates the intended argument is an editorial responsibility.

Accuracy is essential to trustworthy journalism.

Editorial judgment helps make that journalism clear, meaningful, and worth reading.


Graphics created with AI-assisted design tools.
Visual concepts, editorial direction, composition, and final creative decisions were developed by Vallano Media.

From Doing Everything Myself to Directing the Workflow

By Cheval John | Vallano Media
This article was created with the assistance of artificial intelligence.
The final version was reviewed, edited and fact-checked by the author


Graphics created with AI-assisted design tools.
Visual concepts, editorial direction, composition, and final creative decisions were developed by Vallano Media.

For most of my career, writing meant doing nearly everything myself.

Research the story.

Find the sources.

Take the notes.

Organize the information.

Write the draft.

Rewrite the draft.

Check the facts.

Edit the copy.

Come up with the headline.

Prepare the article for publication.

Then start all over again with the next story.

That process was familiar because it was the way I had always worked.

It was also exhausting.

Journalists often wear several hats, especially when they work independently or in smaller organizations.

A reporter may also become the researcher, editor, publisher, social-media producer and, depending on the project, even the person creating graphics.

For a long time, I simply accepted that as part of the job.

Then artificial intelligence changed the way I thought about the work.

Not because AI could suddenly do everything for me.

It could not.

What changed was that I began to realize I no longer had to perform every part of the process in exactly the same way.

I could direct the workflow.

That was a much bigger change than simply learning how to use a chatbot.

The First Stage: Asking Questions

My early use of artificial intelligence was not sophisticated.

Like many people, I started by asking questions.

I treated AI almost like a conversational search tool.

Ask something.

Get an answer.

Ask another question.

Move on.

There was nothing particularly strategic about it.

At that point, I was using AI.

I was not yet working with AI.

That distinction would become important.

The real change began when I stopped thinking only about individual questions and started thinking about the entire process.

What happens before I write?

What happens after I write?

Where do mistakes usually occur?

Which parts take the most time?

Which parts require judgment?

Which parts are repetitive?

Which parts could benefit from another set of eyes?

Once I began thinking that way, AI stopped being only a question-and-answer tool and started becoming part of a workflow.

The Work Did Not Disappear

It is easy to describe AI as a labor-saving technology.

Sometimes it is.

But that description can be misleading.

The work did not disappear.

The work changed.

Instead of spending all of my time manually performing every task, I increasingly spent time deciding what needed to happen, what AI could assist with and what still required my direct attention.

That sounds subtle.

It is not.

Consider research.

A traditional workflow might involve opening a series of tabs, reading articles, taking notes and trying to identify connections manually.

An AI-assisted workflow might help organize documents, compare information or identify questions more quickly.

But somebody still has to decide which sources are trustworthy.

Somebody still has to notice when important context is missing.

Somebody still has to recognize when two sources appear to contradict each other.

And somebody still has to decide what belongs in the final story.

That person is still me.

AI can reduce some forms of manual work while creating another responsibility: supervision.

From Doing Everything to Directing the Workflow

This is the part I did not understand when I first started using AI.

I thought the main benefit would be speed.

Write faster.

Research faster.

Edit faster.

Those benefits are real.

But the deeper change has been learning how to direct the work.

The shift was not from working to supervising.

It was from performing every task manually to deciding which tasks I should still perform, which AI could assist with and where human approval had to remain.

One workflow may be focused on research.

Another may challenge the argument.

Another may examine factual claims.

Another may help transform a completed article into a different format.

Another may help evaluate whether the final product still sounds like me.

Some of those workflows are simple and manually triggered.

Others are more structured.

Not every useful AI workflow needs to be autonomous.

The point is not to remove the human from the process.

The point is to build a process where the human knows when to step in.

I Did Not Start With a Master Plan

The strange part is that this happened gradually.

I did not sit down one morning and announce that I was going to build a collection of AI-assisted workflows.

Much of it came from trial and error.

I encountered a problem.

I tried something.

Sometimes it worked.

Sometimes it did not.

I needed a better way to verify information.

A workflow developed.

I needed a way to challenge weak arguments.

Another workflow developed.

I needed to repurpose an article without losing its meaning.

Another process emerged.

Some workflows worked immediately.

Others had to be changed.

Some became more useful after repeated use.

Some probably were not worth keeping.

Over time, those individual solutions began to form a system.

That is when I started realizing that I was no longer simply using AI.

I was designing ways to work with it.

The Journalist Becomes the Editor of the Output

Traditional journalism has always involved editing.

The difference now is that the journalist may also have to edit what the machine produces.

That means more than correcting grammar.

It means questioning assumptions.

Rejecting weak suggestions.

Checking sources.

Restructuring an answer.

Recognizing when the system has misunderstood the assignment.

Sometimes the most important thing I do is tell the AI:

No.

That is wrong.

That is unsupported.

That does not sound like me.

That is not the question I asked.

Try again.

Those moments matter because they reveal something that gets lost when AI is described as automatic.

The process is often interactive.

The final result may emerge from several rounds of challenge, correction and refinement.

That makes the human role different from traditional writing, but not less important.

In some situations, it can become more demanding because the journalist must supervise both the journalism and the AI-assisted process.

Direction Requires Knowing What Good Looks Like

There is another problem with telling people to simply “use AI.”

You have to know what you are trying to produce.

AI cannot compensate for having no standards.

If I do not know what a strong news story looks like, I cannot reliably judge whether AI has helped create one.

If I do not understand verification, I may not recognize an unsupported claim.

If I do not understand my own voice, I may allow AI to replace it with generic language.

Delegation only works if the journalist still understands the task well enough to judge the result.

That matters because over-delegation carries its own risk.

If journalists stop practicing the underlying skills of research, verification, writing and editing, they may become less capable of recognizing when AI has produced weak work.

AI can increase capability.

It does not automatically provide judgment.

The better I understand journalism, the better I can direct AI inside a journalistic workflow.

A Different Kind of Productivity

Before I began using AI seriously, productivity often meant working harder or working longer.

There were only so many hours available.

If an article took three hours, producing another article often meant finding another three hours.

AI changes that equation, but not as simply as the hype sometimes suggests.

The goal is not to produce as much content as possible.

The goal is to use saved time intelligently.

The most useful question is not:

“How much faster can AI make me?”

It is:

“What should I do with the time AI gives back?”

If AI helps reduce the mechanical parts of a task, I can spend more time checking the work.

I can challenge the argument.

I can examine another source.

I can rethink the headline.

I can decide that a story is not ready.

That is a much healthier definition of productivity than simply publishing more.

Speed only matters if the quality survives.

When the Workflow Fails

Directing AI-assisted work also means accepting that the workflow will sometimes fail.

AI may misunderstand the task.

It may give me information that looks useful but is not.

It may repeat itself.

It may introduce language that sounds polished but weakens the point.

It may send research in the wrong direction.

Those moments are not exceptions to the process.

They are part of the process.

The important questions are simple:

Did I catch the problem?

Do I understand why it happened?

Do I need to change the workflow?

That is how the system improves.

Not because AI becomes perfect.

Because the human process around it becomes better.

Building the System While Using It

One of the unusual things about my experience is that I built many of these workflows before I knew what to call them.

I was solving problems.

I was creating repeatable steps.

I was assigning specific roles to AI.

I was deciding where human approval had to remain.

Only later did I realize that what I had been building resembled structured AI workflows and, in some cases, agent-like systems.

Most of those systems were not fully autonomous.

They were usually manually triggered and specialized.

That distinction matters.

Not every useful AI workflow needs to operate independently.

Sometimes the value comes from building a repeatable process that helps the human work more consistently.

I had not learned the terminology first and then applied it.

I developed the practice first and learned the terminology afterward.

That may be one reason the systems felt natural.

They were built around real problems rather than abstract concepts.

The Managerial Shift

This change also affected how I thought about my role inside Vallano Media.

I was no longer simply writing articles.

I was increasingly managing a publishing system.

Research needed direction.

Verification needed standards.

Drafting needed oversight.

Graphics needed review.

Publishing needed coordination.

AI could assist with many of those functions, but I remained the person deciding how the pieces fit together.

That is where the shift from writer to director became clear.

The word “director” does not mean standing above the work.

It means coordinating it.

The work still has to be understood.

The standards still have to be maintained.

The final decision still has to be made.

But I no longer have to perform every task manually in order to remain in control.

What Other Journalists Can Learn From This

The lesson is not that every journalist needs dozens of workflows.

The lesson is simpler.

Start by examining the work you already do.

Where do you spend the most time?

Where do mistakes occur?

Which tasks are repetitive?

Which tasks require judgment?

Which tasks could be assisted without handing away editorial control?

Then test one small process.

Do not automate the entire newsroom.

Do not start with the highest-risk task.

Pick something manageable.

Maybe it is organizing notes.

Maybe it is generating questions to consider.

Maybe it is comparing two versions of a draft.

Maybe it is asking AI to challenge an argument before publication.

Then examine what happened.

Did it help?

Did it save time?

Did it introduce problems?

What still required human judgment?

That is how a workflow begins.

The Work Is Still Mine

AI has changed how I work.

It has not changed who is responsible for the work.

That was the lesson of Chapter 1.

Chapter 2 adds another layer.

Responsibility does not require doing everything manually.

Control does not mean refusing assistance.

The journalist can delegate parts of the process without delegating accountability.

That is the shift I am still learning to make.

From doing every task myself to directing a workflow.

From asking AI isolated questions to designing repeatable processes.

From thinking only like a writer to thinking more like an editor, manager and publisher.

The technology did not remove me from the process.

It changed where I add the most value.

And increasingly, that value comes from knowing what to direct, what to question, what to verify and when to say no.


Graphics created with AI-assisted design tools.
Visual concepts, editorial direction, composition, and final creative decisions were developed by Vallano Media.

I’m Taking the Nonfiction Writing Challenge Again-This Time With AI

By Cheval John | Vallano Media November 1st, 2026
This article was created with the assistance of artificial intelligence.
The final version was reviewed, edited and fact-checked by the author


Graphics created with AI-assisted design tools.
Visual concepts, editorial direction, composition, and final creative decisions were developed by Vallano Media.

Twelve years ago, I participated in a nonfiction writing challenge.

Generative AI was not part of my writing process.

There was no ChatGPT sitting beside me, no AI assistant helping me organize research and no system I could ask to challenge an argument before I published it.

The process was more traditional: research, notes, writing, rewriting and editing.

Now it is 2026, and I am doing the challenge again.

This time, the blank page has company.

For the next 30 days, I will be working on a new book, Journalist and AI — Volume 1, while documenting what happens when artificial intelligence becomes part of the nonfiction writing process.

This is not an experiment in having AI write a book for me.

It is an experiment in what happens when a writer uses AI as a tool while remaining responsible for the research, reporting, verification, judgment, writing and final product.

That distinction matters.

Twelve Years Later

A lot has changed since 2014.

Writers already had search engines, digital publishing tools, social media and numerous technologies that had transformed how information was gathered and distributed.

But mainstream writers did not have the kind of easily accessible generative AI systems available today.

Now I can use AI to help organize research, examine competing ideas, identify weaknesses in an argument, compare information and assist with editing.

But that does not automatically make the writing better.

It does not automatically make the research accurate.

AI can accelerate good work.

It can also accelerate bad work.

It can help organize information more quickly, but it can also produce convincing errors, flatten a writer’s voice or reinforce a weak assumption if nobody challenges it.

That is part of what I want to explore over the next 30 days.

Why Do This Again?

I have already written books.

So this challenge is not about proving that I can finish one.

I am returning to it because the writing environment has changed so much that I want to experience the difference for myself.

Where does AI actually save time?

Where does it improve the work, and where does it create new problems?

Most importantly, can AI help a writer become more productive without sacrificing accuracy, judgment, transparency or individual voice?

Those questions are part of the reason Journalist and AI exists.

Instead of discussing them only in theory, I am going to put them to the test while writing the book itself.

This Is Not a Perfect Comparison

There is one thing I want to make clear from the beginning.

This is not a controlled experiment between 2014 and 2026.

AI is not the only thing that has changed since 2014.

I have changed too.

I have another 12 years of experience as a writer, publisher and researcher.

Publishing technology has changed.

The projects are different.

The way I organize information has changed.

So if I work faster in 2026 than I did in 2014, I cannot simply say AI caused the entire difference.

AI is an important variable, but it is not the only one.

That limitation is part of the experiment, and I do not intend to hide it.

What I can document is how AI affects the process I use today.

How AI Will Be Used

Artificial intelligence will be part of this project, but it will have defined roles.

I will use it to help organize ideas, analyze information, identify questions, compare sources, examine arguments and assist with editing.

I will also use it as an adversarial partner.

One of the most useful habits I have developed while working with AI is asking a different kind of question.

Instead of only asking, “Is this good?” I often ask:

“What is wrong with this?”

I use AI to red-team ideas and drafts — to search for weak arguments, missing context, unsupported assumptions and alternative interpretations.

But the AI’s criticism is not automatically correct either.

I still have to evaluate it.

The same principle applies to research.

AI can help me find patterns, organize information and work through material faster than I could manually.

But an AI-generated statement is not evidence.

If something matters to the book, it still has to be verified against reliable sources.

What AI Will Not Do

AI will not receive final editorial authority over this project.

I will

I decide what belongs in the book.

I decide whether an argument accurately reflects the evidence.

I decide whether the writing sounds like me.

And if an AI system produces something inaccurate, misleading, unsupported or simply not good enough, I am responsible for catching it.

A system capable of producing polished language quickly can also produce polished mistakes quickly.

Speed cannot become a substitute for judgment.

What Readers Are Going to See

For the next month, I want to make parts of this process visible.

There will be successes.

There will probably be mistakes.

There will be AI suggestions that improve the work and others that I reject.

There may be research that changes the direction of a chapter.

There may be sections that get rewritten completely because the first approach does not work.

And if AI wastes time, gives me weak information, sends me in the wrong direction or produces writing that I throw away, that belongs in the record too.

The goal is not to create the illusion that writing with AI is effortless.

The goal is to document what actually happens.

I will not be publishing every chapter of the book as I write it.

Readers will instead see selected parts of the process, observations from the experiment and what I learn along the way.

At the same time, Vallano Media will continue its regular Tuesday and Thursday publishing schedule, with additional articles published when news, reporting or other projects warrant it.

The book challenge will happen alongside that work rather than replace it.

My Rule for the Next 30 Days

The Write Nonfiction in November challenge does not require me to work on the project every single day.

That is a rule I am adding for myself.

From November 1 through November 30, I will work on Journalist and AI — Volume 1 every day.

Some days may produce a large amount of writing.

Other days may be spent revising, verifying information, reorganizing a chapter or questioning an argument that is not working.

The point is not to chase an arbitrary daily word count.

The point is to keep moving the manuscript forward.

Thirty Days

By November 30, I want more than a word count.

I want a complete first manuscript of Journalist and AI — Volume 1.

But I also want a record of how we got there.

How much did AI accelerate the process?

Where did human judgment remain indispensable?

Did greater speed allow more time for verification and deeper thinking, or did it simply encourage more production?

Did AI help preserve my voice, or did I sometimes have to push back against its tendency to make everything sound polished and generic?

And after 30 days, will I view the relationship between journalists and artificial intelligence differently than I do today?

I do not know all of those answers yet.

That is exactly why I am doing this.

In 2014, I participated in a nonfiction writing challenge using the tools available to me at the time.

Twelve years later, I am returning in a completely different technological environment.

The technology has changed.

The responsibility of the writer has not.

For the next 30 days, you are going to see the process.


Graphics created with AI-assisted design tools.
Visual concepts, editorial direction, composition, and final creative decisions were developed by Vallano Media.