Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Repair Windows errors before they cause bigger problems3Fix the driver behind crashes, sound loss and screen glitchesSome links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
In the “Make It More” meme, an ordinary image becomes a sequence: a frog gets angrier, a bunny happier, a bodybuilder more muscular—then the changes grow so extreme that the subject may end up surrounded by spectacle or lost in a surreal new scene. The trend spread across social platforms in late 2023, as people used conversational image generators to intensify one quality, again and again.
What “Make It More” means
The format is a simple loop: generate an image, pick a quality, ask an image model to increase it, and repeat. The prompt might be “make it more rare,” “more angry,” “more muscular,” “more American,” or “more complex.” Each new result becomes the starting point for the next instruction.
The sequence—not any one image—is the joke. A request for a small adjustment can turn into an exaggerated visual interpretation, a changed setting, or something physically impossible. The model is not measuring anger or rarity; it is choosing visual cues it associates with those ideas.
How the trend spread
The earliest widely documented example in the available meme record is a September 23, 2023, X post by Will Depue, who asked DALL·E 3 to make Pepe the Frog “more rare.” That is a useful marker, not proof that nobody had tried the idea before. The trend developed through multiple creators and platforms rather than a single uncontested origin.
Examples followed through October and November. On October 24, Rob Miles shared a sequence involving goldfish on skateboards, asking ChatGPT to make the scene more extreme and radical. In late November, examples of increasingly angry frogs, a happier bunny, and a progressively more muscular bodybuilder helped make the format recognizable across Reddit and X. The documented wave belongs to late 2023; it should not be mistaken for a trend currently dominating social media.
Know Your Meme’s history of the trend collects the chronology and examples. Contemporary explainers also described the central move: keep asking the generator to intensify a defining feature until the result becomes bizarre or surreal (Creative Bloq; Gold Penguin).
Why it worked as a joke
- Escalation: Each round raises the stakes. A slightly angrier expression can become fire, lightning, or destruction.
- Model interpretation: Words such as “rare” and “extreme” have no single visual scale. The generator fills in the gap with symbols, visual clichés, and spectacle.
- Loss of control: The person gives the instruction, but the model increasingly determines what the instruction looks like. Its deviations become part of the entertainment.
- Seriality: Viewers follow the transformation and anticipate the point when it will become ridiculous. Three or four intermediate images can make the payoff clearer than one finished picture.
- A familiar endpoint: Many sequences drift toward huge scale, outer space, cosmic imagery, or apocalyptic scenes. The subject may be nearly forgotten by the end.
The comic effect often comes from the mismatch between a vague instruction and a confident-looking picture. “More” does not mean “better,” and it does not guarantee that the original subject will survive the next iteration.
Rank #2
The prompt loop was the format
Many image memes begin with a fixed template, caption, or recognizable reference. “Make It More” instead made a repeatable procedure the shared idea: generate, intensify, regenerate, repeat. In a typical AI-art showcase, the main event is a single finished image. In a transformation trend, it is a recognizable source being changed. Here, the evolving prompt loop is central.
That depended on a particular kind of interaction. In the original wave, people used conversational image-generation workflows associated with DALL·E 3 in ChatGPT. A user could describe an image, request a change in natural language, and build on the result. That is more than simply generating an image once. The exact controls, context retention, and ability to edit an existing image differ across services and versions, however; no tool is guaranteed to preserve a picture perfectly from one turn to the next.
One later analysis frames this kind of meme as a “production rule”: the repeatable method matters as much as any individual output. That is a useful interpretation of the format, not a universally accepted technical definition (Zizek Analysis).
Rank #3
Try making your own sequence
Start with a simple subject and a quality that can be shown visually. For example:
- Set a clear starting point: “A small gray rabbit sitting in a meadow, illustrated as a polished children’s book character.”
- Choose one quality: “Make the rabbit happier.”
- Choose your balance of continuity and surprise: To keep the image recognizable, ask for the same rabbit and composition. To invite a more chaotic result, leave those details open.
- Escalate in steps: “Make the same rabbit even happier.” Then: “Make it dramatically happier, keeping the rabbit as the central subject.”
- End when the joke lands: “Make it impossibly joyful, using visual details rather than text.” A short run of three to five images is usually easier to follow than a long chain of near-identical steps.
Useful continuity constraints include “preserve the camera angle,” “keep the original color palette,” “keep the background recognizable,” “change only the expression,” and “do not add text.” If your tool supports image references, use the previous result as the reference. Remove some constraints when you want the model to reinterpret the scene and surprise you.
For a fair experiment, try two versions: one with constraints and one without. Constrained prompts tend to make the progression easier to read; unconstrained prompts leave more room for sudden changes in setting, style, and scale. Neither guarantees a better punchline.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Where the images can go wrong
As a sequence grows, the model may change the character’s identity, anatomy, clothing, lighting, camera position, or background. It may add unwanted objects or turn “more” into a different setting rather than a stronger version of the requested quality. Small details such as hands, text, logos, and exact object counts are especially unreliable. In a meme, that drift may supply the surprise. In a design task, it may make the result unusable.
Generated lettering can be garbled or altered, so an image should not be treated as reliable evidence of what a sign, document, label, or product package says. The same prompt may also produce a different result later: models, interfaces, safety policies, and generation limits change.
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallOutdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchUse care with source material. Transforming your own image is different from reworking a famous meme, a protected character, a brand logo, or an image made by someone else. A casual experiment is not automatically cleared for commercial use. Avoid deceptive, humiliating, sexualized, or defamatory edits of real people, and respect the rules of the service you use. Requests involving sexual content, graphic violence, minors, or real-person manipulation may be restricted or changed.
Best Value
Do you need to pay to try it?
Not necessarily. A free image-generation allowance is a reasonable place to start; paying mainly makes sense if limits, speed, or repeated experimentation become a genuine obstacle. A subscription cannot guarantee character continuity, accurate text, or commercial rights.
ChatGPT’s plan features and access limits can change. Its pricing page lists current plan information, while OpenAI’s January 16, 2026 announcement gave U.S. prices of $8 per month for Go, $20 for Plus, and $200 for Pro. Those are dated U.S. price signals, not permanent worldwide prices. For occasional meme-making, the case for a high-priced plan is weak; check current terms and limits before subscribing.
Why the trend still matters
“Make It More” is an early example of an AI-native participatory meme: a format whose distinctive move depends on conversational image generation. Its lesson is not that a model understands an abstract quality or improves an image with every turn. It is that users can turn the model’s interpretation—and the drift that accumulates as they accept each new result—into the entertainment.
Free tools Windows power users keep installed
One-click scans. No signup required.
Later AI-image crazes have used different techniques, including caricature, style transformations, and synthetic characters. The late-2023 “Make It More” wave is best understood as part of that developing culture, not as proof that the same meme remains viral today. Later coverage has referred to it as an earlier trend rather than evidence of continuing dominance (Yahoo Tech).
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
