Last year, most conversations around AI and writing focused on speed.
Write faster. Publish more frequently. Ship, ship, ship.
And while it’s true that AI can help you write faster, that’s not what matters for many writers who are serious about improving their craft or producing quality work.
Now the conversation has shifted to workflows and thinking, not writing.
Here’s why…
AI can deliver text in seconds, it’s true. But words alone don’t provide meaning and context. That’s your job. And increasingly, readers can recognize cut-and-paste AI at a glance.
AI tools can:
- Mirror tone
- Replicate structure
- Smooth awkward phrasing
- Enforce style and formatting rules
- Help you refine your thinking
AI tools can’t:
- Define why an idea matters
- Understand what’s at stake for the reader
- Create context from lived experience
Those deficiencies offer an advantage for writers.
As researchers at Stanford Human-Centered AI put it, large language models (LLMs) don’t understand the world. They model relationships between words based on probability. In other words, they’re powerful when it comes to identifying patterns, but not so much at grasping and conceiving meaning.
This distinction is important. Writers use words to create meaning from knowledge, research, and lived experiences, yet also from sometimes messy thinking. AI helps sharpen and focus that thinking.
Training AI Trains You to Be a Better Thinker
As a writer, you bring your point of view, your intentions, and expertise to the page. You also subconsciously bring plenty of assumptions. Many of them feel obvious to you, which is exactly where problems start.
What sounds “clear” to you might rely on shared context your reader doesn’t have, an agreement they haven’t granted yet, or leaps they’re not able to make. Basically, they don’t necessarily know what you know. And a skeptical or confused reader doesn’t argue; they bounce.
This is where training your AI tool becomes especially useful. With the right prompt, it can push back on your thinking and expose what you’re taking for granted.
Surface the Assumptions You Didn’t Know You Were Making
Once you start looking for blind spots, assumptions become the obvious place to begin. Assumptions are mental shortcuts shaped by one perspective. And the more familiar you are with a topic, the more invisible those shortcuts become.
One of my favorite uses for AI is to find those assumptions.
I paste my draft into ChatGPT and ask it to name the assumptions I’m making about the reader, the problem, or the context.
It always highlights something I hadn’t considered, such as unstated knowledge or gaps in the reader’s decision process.
For example, imagine you are writing for a client in the education technology space that sells professional training to school districts. You can ask AI to list the likely job titles involved in purchasing decisions, along with the questions each role may bring to the table. Comparing that list to your draft quickly shows where your thinking assumes too much or speaks to only one target audience’s perspective.
Each assumption you surface and address means less ambiguity in your work. And less ambiguity makes your point of view easier to follow, trust, and act on.
Hone Your Point of View Instead of Circling It
I have dozens of half-finished drafts. You probably do too. Every writer has experienced that feeling of writing hundreds or thousands of words, only to realize you’ve lost the thread.
I call it “getting tangled in the words.” This is usually not a writing problem but a thinking problem. My point of view wasn’t clear enough from the outset, and I kept trying to find it by moving my fingers across a keyboard. This method can work eventually, but AI can help you get there faster.
AI can help you re-anchor your thinking before you spend hours writing in circles.
I’ll use a prompt like:
“This is my messy draft around [topic]. I’ve wandered from my primary point. Can you pull out my strongest point of view and tie it to my ideal reader?”
Then, depending on what comes back, I might follow up by asking what’s strongest and what needs sharpening. In many cases, that’s all I need to get back on track.
Ask AI to Explain Your Idea Back to You
Some ideas make sense intuitively but fall apart when you try to explain them. That’s not a failure. It’s useful information, and it’s where AI works well as a clarity test.
Ask AI to summarize your argument, then look at the response through three questions:
- What problem does this solve?
- Who is it for?
- What decision does it help the reader make?
If the answers come back vague, you’re not ready to revise and polish. You’re still in thinking mode, and that’s where good writing starts.
Use it to Stress-Test Your Thinking
Once your core idea is solid, you can use AI to challenge your draft and improve it.
Try questions like:
- What might a skeptical reader push back on?
- What feels under-supported?
- Where does it need improvement?
AI doesn’t replace your judgment. It highlights weak spots so you can strengthen your thinking and produce a stronger piece of writing.
Let AI Enforce Rules
AI is also excellent at guardrails:
- Tone of voice
- Brand language
- Structural consistency
When you’re juggling five or six clients, each with their own preferences, those details take you more time and can drain your energy with too much to remember. Do you use the Oxford comma or not? Are the subheadings title case or sentence case? What are the preferred terms?
AI excels at ensuring you follow each client’s guidelines. You can focus on the writing. This separation can free up space for your thinking.
The Real Shift Isn’t Speed
The writers who get the most value from AI are those learning to use it to think more clearly.
AI is a mirror. It reflects ambiguity and exposes patterns. That feedback loop helps you catch weak thinking while it’s still fixable.
Used this way, AI doesn’t replace your point of view or dilute your voice. It sharpens your thinking and strengthens your writing.
How are you using AI tools?