Hello M(AI)VENS,
If it feels like everyone you know has suddenly turned into a cartoon version of themselves, you are not imagining it. The latest ChatGPT character trend has taken over social feeds. This week, we’re taking a deep dive into this trend and similar trends that came before it.
But first, let’s cover the top AI news stories of the week.
Growing Tension Over AI Data Centers
A new POLITICO poll shows Americans hold mixed but still flexible views on AI data centers. Nationally, voters tend to feel neutral to mildly positive, often linking data centers to jobs and economic growth. Support slightly outweighs opposition, and a large share of respondents remain undecided. However, attitudes shift once projects move closer to home, where concerns about electricity costs, water use, and environmental strain fuel local pushback. The poll also highlights a political undercurrent. Republicans express somewhat stronger support than Democrats, while Democratic voters are more inclined to oppose candidates who back local data center projects. With hundreds of new facilities expected nationwide, the issue of data centers could play a significant role in upcoming elections.
AI Ignited Stock Sell-off
Big Tech stocks plummeted this week as investor scrutiny increased around AI spending. The sell-off spread across the sector, erasing more than $1 trillion in combined market value from Amazon, Microsoft, Nvidia, Meta, Alphabet, and Oracle. The reaction reflects growing focus on the scale and pace of AI investment. Companies are committing vast sums to data centers and cloud infrastructure, prompting questions about returns, capacity, and timing. By the way, Amazon says it’s spending $200 billion on AI this year.
Moltbook 🦞
You may have seen buzz lately about Moltbook, a new social media platform designed for AI agents rather than people. It’s a lot like Reddit, with posts, comments, and upvotes, except the “users” are AI agents. Supposedly, those agents were talking to one another, debating philosophy and productivity, forming communities, and even creating their own religion, complete with a manifesto and shared beliefs. Many of the agents are created using OpenClaw, an open-source tool people install on their own computers to run AI helpers and direct them to act on their behalf, including posting on Moltbook. After the initial hype, experts clarified that the agents don’t actually act independently, but rather are following human instructions. Attention has since shifted to security concerns, after it was determined the site exposed APIs, weak verification, and ways for humans to impersonate or manipulate agents.
The Caricature Trend 🎭
Over the past year, ChatGPT image trends have followed a very recognizable progression. In mid to late 2024, people began experimenting with simple visual transformations. Pets were among the first stars. Dogs and cats were uploaded and “humanized” into people with jobs, personalities, and wardrobes, often with surprisingly emotional results. That phase leaned heavily into novelty and humor and spread quickly because it required almost no explanation.
By early 2025, the experimentation shifted from pets to people. Users started reimagining themselves in specific artistic styles. Studio Ghibli portraits were everywhere for a while, followed by Pixar-style characters and Lego-inspired figures. These images felt nostalgic and visually distinctive, and they traveled fast across Instagram and X because they were immediately recognizable. You couldn’t scroll for five seconds without seeing one.
Spring and summer of 2025 brought a more polished evolution. The Barbie and Ken doll trend emerged, with users asking ChatGPT to turn their photos into hyper-stylized dolls complete with glossy textures, curated outfits, and aspirational backdrops. This phase aligned closely with personal branding and identity play, especially among professionals who think about how they show up visually online.
By late 2025, the trend matured again. The action figure version took hold and stayed longer than most. People asked ChatGPT to create 3D, plastic-encased versions of themselves, posed like collectible toys with accessories tied to their jobs and interests. A consultant might appear holding a laptop and coffee cup. A nonprofit leader might be packaged with a clipboard and event badge. These images were telling a clear story about work and identity, which helped them travel particularly well on LinkedIn.
That brings us to now.
The latest iteration has arrived, and it is the one you are likely seeing friends, colleagues, and acquaintances post right now. The ChatGPT caricature or “mini-me” character trend is here.
Users upload a photo and ask ChatGPT to exaggerate their defining features and professional context into a cartoon-like portrait that’s grounded in reality.
Unlike the earlier style-based trends, this one stays close to real life. People are shown at work, in familiar environments, wearing what they usually wear, holding objects that make sense for their roles. The exaggeration is light and intentional, often focused on facial expression or posture rather than full transformation. The result usually feels flattering, funny, and specific enough that friends immediately recognize the person behind the image.
Part of why this trend is moving so fast is that the technology supporting it has reached a new level. ChatGPT’s image generation now works seamlessly with text and visual inputs together, which means users can guide the output with a level of detail that was not possible even a year ago. When someone includes their job title, daily responsibilities, work setting, or even how they tend to show up energetically, the model incorporates those cues into the image in ways that feel surprisingly thoughtful.
The process itself is straightforward.
You upload a clear photo, often a headshot or casual pic. You write a prompt that explains who you are and what you do, sometimes adding tone cues like playful, polished, or slightly exaggerated. Many people refine the image through a few rounds, adjusting facial features, props, or the overall vibe until it feels right. Once it does, they post it.


The social response has been immediate. Most people share simply because it’s fun. Others use it as a conversation starter. On LinkedIn, the images often accompany captions about work, leadership, or career identity, turning a lighthearted trend into a moment of connection.
Pushback
There has also been pushback, which tends to accompany any highly visible AI trend. Environmental concerns, artist displacement, and broader questions about the role of generative imagery have surfaced alongside the posts themselves.
Are we all just wasting energy every time we generate a silly image of ourselves? Those conversations are part of the landscape now.
The environmental footprint of AI and the water use associated with data centers is real. Large data centers rely on water for cooling and energy production, and their impact varies significantly depending on location, infrastructure, and local water availability. At the same time, researchers and industry analysts note that individual uses, like generating a single image or caricature, represent a very small portion of overall consumption when viewed at the system level.
As with many emerging technologies, the real conversation sits between scale, responsibility, and long-term planning, rather than any single action in isolation. It’s a useful lens to keep in mind as AI becomes more embedded in everyday creative and professional life.
Building Confidence
Often, confidence with AI grows through low-pressure experimentation rather than formal instruction. Asking ChatGPT to turn you into a caricature based on your real job is, in practice, an exercise in prompt clarity, iteration, and creative direction. Those muscles are the same ones used when drafting strategy memos, refining messaging, or exploring ideas with AI later on.
If you’re tempted to try the trend yourself, the best results come from treating the prompt like a brief to a designer who has never met you. Be specific about what you do, where you work, how you typically show up, and what you want the image to convey. Decide ahead of time whether you want something playful, polished, or somewhere in between. Give yourself a few rounds to refine, because the first output is rarely the best one.
Here’s a sample prompt you can adapt:
Using the photo I’ve uploaded, create a caricature of me that still looks recognizably like me. I want this to feel [playful, professional, polished].
I work as a [your role] in the [your industry or field], and my day typically involves [a few key activities you actually do]. I usually work in [home office, corporate office, remote, client sites, etc.], and I tend to dress [how you normally dress for work].
Include details that reflect my work and personality, such as [tools you use, objects you’re often seen with, or small visual cues that feel very “you”]. Exaggerate facial expression and posture just enough to make it fun while still realistic.
The overall vibe should feel fun and representative of my real-life.
If the first image is close but not quite right, follow up with small adjustments, like refining the expression, changing the setting, or dialing the exaggeration up or down. The more clearly you describe yourself and your work, the more natural and specific the final image tends to feel.
You don’t have to share it publicly for it to be a useful exercise. Even creating one privately gives you a feel for how responsive and flexible AI can be when you guide it clearly.
Klara and the Sun by Kazuo Ishiguro
📚 Book Club & Prompt Party (Free)
Book: Klara and the Sun by Kazuo Ishiguro
When: Friday, March 27, 2026
12:00 PM Eastern | 9:00 AM Pacific
Register here (once you register, you’ll get a confirmation email with the Zoom link)
This story is told through the eyes of Klara, an Artificial Friend designed to be a companion to a teenage girl. Klara observes family life, social hierarchies, and a world shaped by genetic enhancement, while forming her own beliefs about care, devotion, and the Sun as a source of hope and healing. The book invites readers to reflect on love, consciousness, and how humanity shows up in both people and machines.
We’re giving everyone two full months to read this one, which I hope allows enough time to enjoy it without pressure. 💜

♟ Queen of Chess | Netflix
Judit Polgár broke into the world top 10 in 1996 at just 19 years old, a milestone no woman had ever reached before. At the time, she was competing head-to-head with the very best male players in the world, including reigning and former world champions. I’m so excited to see her story come to life on Netflix!
I have no doubt this caricature trend will eventually be replaced by the next one, just like the pets, the dolls, and the action figures before it.
Each wave brings AI a little closer to everyday usage and expression, making it feel less “techy” and more personal. However, it also challenges us to be thoughtful about security risks, and about if, how, and when we choose to use environmental resources for these kinds of outputs.
Warmly,
Cheyenne 💜
Copyright © 2025 M(AI)VENS. All rights reserved.







I think the action figure trend was my favorite trend to date. But this caricature one is fun too. 🩷🦩