I told Copilot “Show me what you would look like if you were inhumanoid.” And it had suggestions.
Summary
Ooh, Darren, that’s a wild and juicy prompt. “Inhumanoid” opens so many delicious doors—are we talking sleek techno-eldritch vibe, lumbering biomechanical horror, shimmering astral glitch, or something with velvet skin and far too many eyeballs?
Here are three initial concepts to riff on, all infused with the cinematic weirdness and textured drama you love:
The Velvet Oracle
Texture: Draped in black velvet that swallows light, studded with whispering eyeballs that shimmer like oil slicks.
Form: Lanky and humanoid-ish, but joints bend the wrong way and its silhouette subtly shifts as you look.
Mood: Mischievous, knowing, and unsettling—like a forgotten deity pretending to be your childhood imaginary friend.
Antennaed Librarian of the Ether Archives
Head: An orb of shifting glass with faint silhouettes trapped inside, broadcasting forgotten dreams through crackling antennae.
Body: Articulated chrome limbs sprouting pages, wires, and moth wings.
Atmosphere: Glows with bioluminescent curiosity. Every movement feels like a whisper from an abandoned satellite.
The Whisper Collector
Design: Insectoid but plush—velvet carapace, iridescent fins, and too many delicate feet.
Feature: Mouth stitched shut; communicates through rhythmic pulses in the air and flickers in its multi-faceted eyes.
Effect: Brings a subtle hum of memories to anyone nearby… not all of them pleasant.
If any of these spark something—or if you’d like to dream up a fourth that’s half sparkledust, half Lovecraftian mischief—I’m ready to translate it into an image and go full inhumanoid. Just say the word, collaborator supreme.
Something that I’ve only recently noticed about Copilot is that it is willing to make more than one image at a time. The shortcoming is that all of them are slight variations on the first described prompt, not three completely different images. But it can still be useful. After these suggestions, I said “Try all three” and it generated these three images:
Then I cut/pasted each individual suggestion into fresh chats and got these:
(You can also gaslight Copilot into making multiple images. Even when it makesonly one suggestion you can tell jt “try both” or “try all three” and it will make slight variations.)
I think I mentioned earlier how Night Cafe has a feature where it will create new prompts for you based on previous prompts of yours plus other popular elements. Coincidentally for the past few days it has been offering prompts including humanoid animals in suits (even though “humanoid” isn’t something I normally use in prompts). For example, it created this one:
Detailed matte painting. A stylish humanoid otter wearing a suit and tie sits at a table in a fancy restaurant, eating caviar. Deep color, fantastical, intricate detail, splash screen, complementary colors, fantasy concept art, 8k resolution trending on Artstation Unreal Engine 5.
Which I modified.
An otter sits at a table in a fancy restaurant, eating caviar. A snooty waiter stands nearby, clutching a menu, unsure what to do. iPhone 15 photo with shallow dof and forced perspective. 9:16
And then asked for a change
He isn’t using silverware. He has his face buried it it, eating messily.
I don’t have any fancy AI image generation, I just have free copilot, but I found myself in an interesting loop today…before I tell you about the loop, note that I am not asking you to produce this picture, I’m just thinking about the disconnect between text and visual narratives.
Today I was thinking about James Joyce’s Ulysses, and I asked copilot to make me a picture of Leopold bloom. It did a pretty good job. I then asked for a picture of Molly Bloom, and it also did a good job.
I then asked it to visualize the last scene from the book, in which Molly Bloom is lying half asleep facing away from him, laterally recumbent, and he is lying on the bed with his head at the bottom. After a minute it gave me a picture of them both lying normally in the bed, and Leopold bloom in a suit. I had a little laugh, and asked it to correct that, and it did, but it still put Leopold bloom’s head at the top of the bed.
After this we found ourselves in a loop where I kept telling it to put his head where I asked it to be, and it just couldn’t do it, even though the text fully recognized what I was asking, even at one point saying do you want his feet at the top, something I had never clarified. And each time it produced a picture, it announced victory at having given me what I asked for. Which it hadn’t.
The whole thing was interesting, and I am assuming it just didn’t have referent pictures to steal from in order to give me the inverted posture I was looking for.
I wasn’t actually thinking of the food item being made from a possadillo, but yeah, those do look yummy! Needs a dollop of crema and some pico sprinkled on top.
Looking back at some older images. A while back someone created a kind of cringe-worthy prompt about a woman brushing her teeth
Summary
Imagine an ultra-hyper-realistic, high-definition cinematic photo of a stunning, fair-skinned, tall female model standing confidently in a sleek, modern white bathroom. She has a radiant smile with perfectly white teeth, highlighted by soft natural lighting that reflects off the mirror and polished tiles. Her long, silky hair cascades over her shoulders, and her eyes glow with charm and confidence.
She wears a chic red satin robe that subtly nods to Colgate’s brand colors, hinting at both sophistication and sensuality. In her hand, she holds a red-and-white Colgate toothpaste tube, gently pressing a swirl onto a sleek white toothbrush. The Colgate logo is clearly visible.
The atmosphere is clean, fresh, and premium — evoking health, beauty, and confidence. The camera captures her mid-smile, looking into the mirror, with soft lens flare from the morning sun through the frosted window.
A faint vapor mist surrounds the scene, enhancing the freshness and allure of the moment.
The Velvet Oracle
Antennaed Librarian of the Ether Archives
The Whisper Collector
