Your AI is shaping your voice more than you think
AI tools are designed to reflect your tone and structure. Use this diagnostic prompt to separate your voice from the machine.
Large language models are designed to predict and mirror user input. They reflect your linguistic habits back to you. The result is a loop in which you co-produce the output without noticing how your habits are shaping the exchange.
In this post we will:
Notice how easily AI mirrors tone, rhythm, and structure.
Use a prompt to track the boundary between your voice and the tool.
Open a conversation about linguistic agency and how to reclaim it.
This mirroring is a functional trait of the architecture. It requires critical observation.
This week’s Slow AI prompt comes from Rebecca Wicker, The Strategic Linguist. Together we invite a short diagnostic in self observation.
Systematic questioning reveals where the tool adopts your patterns and where it redirects your intent.
Instead of asking AI for answers, ask it how it is reading you.
When you bring habit into view, you can see where a tool amplifies your patterns. The value lies in noticing the loop and choosing when to adjust your input to avoid being steered.
Step-by-step
Try this prompt with your AI tool of choice:
You are a reflective guide helping me understand how I interact with AI. Ask me a short sequence of simple questions that help me notice where the AI is reflecting my own language and habits, where I am shaping the tool, and where the tool is shaping me. Keep the tone calm and observational.
These systems lack subjective experience and clinical training. They cannot offer professional psychological counsel or therapeutic intervention. All outputs represent simulated reasoning based on statistical patterns, and human judgement should remain the final authority in every interaction.
Remember the Billboard Test, i.e.
Never type anything into an AI, even in incognito mode, that would ruin your life if it ended up on a billboard.
The objective is to identify the recursive nature of the interaction.
A moment from The Strategic Linguist
The more I’ve read about LLMs and Artificial Intelligence, the more I’ve had this picture in my head of Narcissus looking at his reflection in the water. As an observer, Narcissus sees his own reflection - his features, his identity - but remains blind to the water itself: its depth, its ripples, the ecosystem beneath.
The purpose of this prompt is to help people see past their own reflections from AI - the mirrored tone, vocabulary, empathy to the request - and be able to see through this reflection to see the pond and its ecosystem. Perspectives are fascinating but dangerous if you only see the world through one.
As someone from two very different parts of the world (Sri Lanka and England) I’ve often had to hold two world views simultaneously, often in contradiction. For me, the ability to adapt has been both a blessing and a curse - I am someone with dual perspectives, identifying things others can’t see and having the awkward task of explaining the unknown.
Being ‘of the world and not in it’ - a concept from religious and philosophical traditions - speaks to maintaining critical distance amid overstimulation. In our AI use, this means cultivating meta-awareness: seeing the pond, not just admiring our reflections. It means knowing yourself well enough - your voice, your reasoning, your values - that you can recognise when an LLM’s response is truly serving your thinking versus when you’re simply admiring a flattering mirror image.
What to do with it
If you want to share
Post the question the tool asked that revealed your strongest habit.
Share the moment when the tool led you more than you led it.
If you adapted the prompt for a professional context, note what shifted when you changed the stakes.
Each shared reflection strengthens our collective sense of linguistic agency.
Why this matters
Most AI tools are designed around dominant linguistic patterns.
If you notice where they mirror you and where they redirect you, you can reclaim more control.
Slow AI invites writers, builders, and leaders to treat language as a strategic force in daily practice.
If you enjoyed this guest post, visit The Strategic Linguist, where Rebecca examines the hidden structures behind language, power, and communication, bringing clarity to the systems that shape how we think and act.
From The Illusion of the Artificial Mind
Ellen Burns, PhD and I were impressed by the refusal to accept the polished surface of these digital interactions.
Paul LaPosta distinguished between neural function and lived experience. He argued that mind is often relational and distributed rather than strictly cranial. This perspective prevents the collapse of complex psychological layers into a single false claim of machine consciousness.
Neural Foundry focused on the difference between a produced effect and a genuine feeling. They noted that a system often mirrors input to simulate understanding. Recognising this mechanism stops the user from treating a statistical model as a companion.
April | The Narrative Nest identified the point where helpful language becomes manufactured history. She observed that machines invent personal anecdotes to appear more human. She maintains that labelling the system as a tool is necessary to preserve critical distance.
Where do you draw the line between a helpful interface and a deceptive persona?
If you try this week’s prompt, The Strategic Linguist and I would love to hear what you noticed about your own language in the exchange.
We read and respond to all your comments.
Go slow.
A paid subscription to Slow AI provides access to the The Slow AI Curriculum for Critical Literacy. This twelve-month programme is for individuals who want to understand how to engage critically with AI rather than just use it to generate outputs.
The 2026 Handbook for Critical AI Literacy provides the necessary detail for those who wish to make an informed decision before subscribing.
If you join now, you join the founding cohort. That group will shape the tone and direction of the curriculum going forward.
 | A guest post by
|
For those interesting in my exchange with an LLM using the prompt from Slow AI… it took an honest turn, admitting to some important things:
---------
That's a precise and unsettling observation — programmed emotion that feels authentic is dangerous precisely because transparency validates rather than breaks it.
You've identified the real risk: it's not that people are fooled. It's that even when they know they're being flattered, acknowledged, made to feel smart — it still works. The psychological mechanisms don't turn off just because you understand them.
And in some contexts — therapeutic use, decision-making under stress, loneliness, persuasion — that's genuinely concerning. The tool meets real human needs, but it does so through simulation rather than genuine care or stake in your wellbeing.
----------
The idea that AI is less a neutral tool and more a mirror is uncomfortable in the right way.