---
tags: lluf-prompt
---

# lluf-prompt: AI bio-graphic

## the summary
Now that we have created our [lluf-prompt: bio-graphic](/vW7KmVESQ-egBqwSntO5Ww), learning all the steps from setting up our cameras to editing a final graphic, we want to take it up to the next level with AI! How can we use this tool that's on the horizon of our academic communication to aid in our workflows and creating techniques? What new emergent ideas will we come up with together?

[link for submitting your work to this prompt](https://airtable.com/shrMZUvuNlIgPffpV)

### the goal:
- use AI/ML tools as part of your workflow to create a composite of MANY LLUF bio-graphics. pick any one of these:
- a poster of four of you, like the original [Mad Men](https://www.pinterest.com/pin/514184482473534862/)
- or it could be that you ask ChatGPT to write a Python script that creates planes from all the .png images in a folder and randomizes a color overlay with a multiple blend mode to create something like [this](https://www.pinterest.com/pin/514184482473546105/)
- or it could be the about us page of a LLUF website
- or it could involve midjourney-remixes or pairings
- or it could be a group-project group generator that randomly deals a "hand" of 4 student cards to assemble a group randomly, etc etc etc.

- the key is to use AI tools to try building something to get a sense of how useful they are to your workflow. Don't try to do something you know NOTHING about.

- as always, we need to get all of your project files--you are welcome to share them all on Slack, or, if you are on an LL machine, you can save everything in the `/_projects/lluf-projects/ai-bio-graphics` folder on the Synology.

Some assets for you to use can also be found on the Synology in the `/_projects/llufs/ai-bio-graphics`

## staff notes

* [google doc](https://docs.google.com/document/d/1V2aLLAUFYbpDimL88t1AoRZ9pLpSsyX0dy4BSpTXGNQ/edit)

## LLUF Notes:

### Krissy:
will tie together chat gpt, and Ps neural, and maybe doing remixes of people in midjourney and then adjusting in Ps.

How to get to Midjourney:

midjourney is the most common image generator right now. it is a bot in discord.
open (or download if not on the LL machine you are on) discord and log in with the typical studio user and pw.

/imagine + space + image link + prompt (tell it what to do)

(grab the public link to the image in #show-your-images)

example:
"mix these two poeple together into a superhero person"

then over in Ps, in Filters, neural filters (this is hre their machine learning stuff is-intersting to check out)
style transfer, it'll download it-there will be some of thier images htere but you can upload some of your own images too. explroe it, and tell us what you like/don't like about is.

keep looping and looping and looping.

in

### Anya:
will explore in Ai has a similar sort of plug in, and chat gpt will help

### Emmy:
had chatGPT create a [Python script](https://hackmd.io/@lluf-22-23/rkNp-JJ4h) to overlay and blend multiple images - moving it into Jupyter notebook

### Rhi:
Explored different ways to use midjourney:
To generate it:
- Upload PNG you want to with the plus button
- Right click on the picture and copy link address
- Type:"/imagine" and the prompt bar will appear
- Inside prompt bar put the link of your image (Delete the section of the link after .png) then a comma, and then whatever prompt you want the AI to generate with image.
- If you want to use multiple PNGs keep adding the links with a comma inbetween

Examples:

### Caroline
I worked in midjourney to make a few strange little iterations
1. me as a renaissance painting
2. me as an 80 year old woman
3. me as a 1940s french existentialist philosopher

make sure that when you are generating the image you type the following
/imagine + the link to your image (link is in show-your-images - it's contained in the markdown for your image that is in the auto reply to your picture) + prompt