When I saw the trend recently of some of my online friends posting an illustration of themselves created by ChatGPT which encapsulates their “vibe”, I was curious to see what it would create for me. I wrote this post to capture my own musings and observations more than anything else.
This is the prompt that was shared along with two photos (I chose two which were available publicly on my Instagram account):
Draw the character in the image, (Insert name) in a free and stripped sketch style. On a bright white background, freely distribute full-body drawings, face close-ups, small scribbles, full-body sketches, and even chibi/deformed versions, so that the page conveys the character’s humor and personality. Don’t do it like an organized character sheet, but like a sketchbook full of information drawn at will by an illustrator and then stacked.
Use everything you know about me from our conversations, including my personality, habits, strengths, quirks, profession, and overall vibe, to imagine how an illustrator would interpret me as a character.
Let me preface the next part by sharing one of my go-to prompt stems: CAST (from an ASCD publication)
Criteria-constraints, rules, norms to follow
Audience-who the output is geared towards
Specifications-specific details, examples
Test-revise, reflect, refine
The “test” piece is the reason I like this model so much because I always want to emphasize the necessity of a human checking an output and we know that learning comes from reflection. When I talk about the test phase, I also invite people to use the same prompt on a different platform to question the result.
In the case of this prompt, C(criteria), A(audience), S(specifications) were already embedded. But I was curious about what other platforms would create for me and WHY. I began with Chat GPT because I use this platform most frequently and this seemed to be part of the directions.
This was the first iteration and at first glance, I thought, Wow. So much of this really does embody who I am. The dress is spot on and so is much of the messaging in terms of my teaching philosophy. But with AI, a first glance should never be the last glance!
I then looked again and thought about how many of the details were just made up (as AI is apt to do). Even on my best make up days, I don’t look that good! I am a girl mom, but the depiction of the teens in the image is boys. I do love coffee, but no where in my previous prompts did I ever mention coffee. I was also shocked to see an image of me on a hike (which only just happened!). I had forgotten that I had asked for advice on my Camino journey when I was in the planning stages. I question why it selected that detail and not the several recipe ideas I prompted or the beach day ideas I asked it to brainstorm? I also don’t work specifically with multi-language learners although this was a lesson within my teacher education class and noticed the reference to languages. And so, curious, I followed up with, “tell me more about some of the details included in this sketch”. This is the output I received:
Absolutely. The sketch wasn’t based only on the two photos you uploaded—it was a visual interpretation of the picture that emerges from our conversations over time. An illustrator creating a “character sketchbook” page would typically exaggerate recurring themes, values, and quirks rather than just draw physical features.
Here are some of the details and where they came from:
The central figure standing confidently in front of audiences came from:
- Your frequent work with educators, teacher librarians, and school leaders
- The presentations you’ve discussed on AI, digital citizenship, media literacy, and digital wellness
- The conference photo you shared from Queen’s Conference on Education
The Educator and Speaker
The pose is meant to communicate someone who spends a lot of time facilitating conversations and inspiring others.
The Digital Leadership References
The notes about:
- Digital leadership, empathy, AI and ethics, media literacy, cyberbullying and well being, came directly from recurring topics in our conversations, especially:
- Your work around digital citizenship
- Your presentations for parents
- Your focus on helping young people navigate technology wisely rather than fearfully
- Your interest in cyberbullying, sextortion, and online safety
“Technology Should Connect, Not Harm”
This wasn’t a quote you provided. It was an illustrator-style synthesis of your overall stance.
Across many conversations, you’ve consistently emphasized:
- Human flourishing
- Empathy
- Relationships
- Digital wellness
- Using technology thoughtfully rather than rejecting it
So the sketch turned that philosophy into a short visual slogan.
That came from a combination of:
The Chibi “Internet Safety Superhero”
- Your public advocacy work
- Your parent workshops
- Your role helping adults and youth navigate online challenges
An illustrator often turns a person’s mission into a playful superhero version of them.
The Multilingual Learner References
The globe and references to language accessibility came from:
- Your work with MLL/ELL learners
- Your presentations about AI supporting multilingual learners
- Your interest in inclusion and equitable access
The Library and Books
The books, notebooks, and learning spaces came from:
- Your work with school libraries and teacher librarians
- Your research into Library Learning Commons
- Your focus on inquiry and information literacy
The Camino Sketch
The backpack, walking stick, and Camino sign came from:
- Your planned Portuguese Camino walk from Vigo to Santiago de Compostela in 2026
An illustrator would likely see that as a metaphor for lifelong learning and personal growth, which is why it appears alongside reflective notes.
The Parent and Teen Conversations
The scenes showing you talking with students and parents came from:
- Your parent workshops
- Your work helping families discuss technology
- Your recurring focus on empathy and conversation rather than control
The Coffee Mug
The mug wasn’t based on anything specific you’ve told me. That’s a common visual shorthand illustrators use for:
- educators
- writers
- speakers
- reflective thinkers
It was an inferred personality detail rather than something I knew.
The Curious Part: What Was Invented?
A few things were creative extrapolations rather than facts:
- The exact slogans
- The superhero cape
- The specific notebook pages
- The coffee mug sayings
- Some of the decorative doodles and checklists
These were attempts to capture your vibe rather than represent actual biographical details.
If an illustrator were creating a second version based even more closely on our conversations, I suspect it would lean even harder into:
- teacher librarianship, AI literacy, digital media literacy, kindness and empathy, inquiry-based learning, parent engagement, conference speaking, books and reading, your “human-centered technology” philosophy and a little less into generic educator imagery. The resulting page would probably feel like a cross between a sketchnote, a conference keynote, a library learning commons, and a digital citizenship workshop.
And so I asked it to create a second version and this is what it created:
For some people an output like the one above which is so close to how I see myself and such a positive summation seems both magical and gives the illusion that ChatGPT somehow “knows” you. Asking myself what’s missing and why, what was added and why, is a good exercise in AI literacy. AI is a prediction machine not a knowledge machine. It generates responses by predicting patterns in language, which can produce remarkably useful outputs but can also lead to errors, bias, or fabricated information. Thus the “leaps” it took: for example using my Camino walk to infer that I was a life-long learner and the slogan, “We don’t need perfect, we need connection” which is actually not something I ever said (although I do like it!)
I then went into Google Gemini (with my three different platforms) and interestingly, the first time I included the prompt, Gemini (in my District account), would not create the image. It said, “There are a lot of people I can help with, but I can’t depict some public figures. Do you have anyone else in mind?” I then had to proceed to give consent. Fascinating. I didn’t have the same issue when I tried it in my university account because I added (me) after my name. I need to look into this further. The results here? General and more concretely connected to the content of the images I uploaded. The outputs here were not as reflective of the tasks I asked it to create in chat history and I wonder if this is a safety mechanism within District accounts?
Gemini on my K-12 District account:
Telescopes? The conference I keynoted for a Faculty of Education (from the image) used the theme, “Teaching Beyond the Stars: Charting the Future of Education”. Science is NOT in my wheelhouse. At least I look more my age in this one, but what is in my hair and why? Why am I taking a selfie with aliens? I also find the October 23rd date so very arbitrary!
Gemini on my Faculty of Education account:
This sketch attempted to generalize my personality. Again with the aliens and telescopes. When all else fails, I suppose repeating a statement with a slightly different image will take up more space?
I followed up with Co-pilot, which I use very infrequently. The resulting image had elements which were actually non-sensical which was surprising but also not given my lack of chat history. Again-coffee! And if I am a presenter, why is there a male figure presenting about tech?
We would never want students to upload images of themselves, but I wonder if showing kids our graphic might be a great conversation starter to support the importance of critical thinking skills, the necessity of always questioning the output we get. It is easy for kids to think of AI as a “friend”; a fact that is emphasized by the response “the picture that emerges from our conversations over time” an important myth to dispel when talking about what AI is. The questions, “What’s missing?” “Why am I getting this output?” are also really important questions to emphasize.
If you tried the prompt, would love to hear what it got right and what it got wrong!
