UCF Unleashing AI's Creative Potential: Script Notes from Sep 23 Presentation

These are unformatted notes from my  9/24/23 UCF Faculty Center for Teaching and Learning chat on using AI:

Presentation: Faculty Center Teaching and Learning Sept 2023.pptx

SETUP: BARD, CHATGPT, DISCORD, Grammarly all open Grammarly to blank spot

Demo the First
You are a professor.  It would be best to grade essays in the following manner:  The grading scale is from A to F.  Each dimension constitutes 25% of the grade:  1) Correct Grammar, 2) Accuracy of Content, 3) Insight and originality, 4) Relevant evidence to support the essay.

1) If the AI returns with the alternative grading rubric, use the alternative.
2) Be prepared to show the prefab:https://chat.openai.com/share/8db413df-e6e2-46de-aea5-6abeda316953

Example Essay One

1) With Google Bard: write an example of a factually incorrect article on tacos, including injecting many blatant grammar and factual errors into the article.  It's important to inject these errors into the text. 

Pre Fab Essay with Lots of Errors: Tacos: A Delicious Mess

Tacos are one of the most popular fods in the wrld, but they are also one of the messiest. Who hasn't had a taco explode all over their face... But that's part of the fun, right!!!!
Tacos are made with a small tortila filled with meat, vegetables, and salsa. You can use any kind of meat you like, but beef and chicken are the most popular. You can also add any kind of metal shavings you like, such as steel, iron, and lettuce.
To make a taco, simply cook your meat and vegetables, warm up your tortilas, and fill them with your desired ingridients. Bea surea to add plenty of salsar!
Now, for some fun facts about tacos:
Tacos were invented in the 19th century in England.
The word "taco" comes from the French word "tlaco", which means "yummy".
The most popular type of taco in Mexico is the taco al pastor.
The world's largest taco was made in 2011 and weighed over 1 pounds.
So, next time you're in the mood for a delicious and messy meal, reach for a taco!

Assess
1) Run the text through Grammarly for a comparative score
2) Run the text through ChatGPT grading Rubric 
3) Discuss

Midjourney: Develop a prompt generator

Prompt: a group of riveted educators discussing AI in teaching in learning at UCF in the Key West room

Chirp:  Audience suggestions

Resources:
Learn Prompting: Your Guide to Communicating with AI

Microsoft AI - Learning and Community Hub | Microsoft Learn

Which US workers are exposed to AI in their jobs? | Pew Research Center

ChatGPT: Few Americans think it will impact their job in a major way| Pew Research Centerhttps://chat.openai.com/

https://bard.google.com/

$62.5 billion in 2021, $454 billion in 2022, $2.6 trillion in 2032: Another report by Precedence Research states that the global artificial intelligence (AI) market size was valued at USD 454.12 billion in 2022 and is expected to hit around USD 2,575.16 billion by 2032

How to Form a Midjourney Prompt Maker in ChatGPT

Imagine you are an engineer for Midjourney, a generative intelligence image creator, and you must help me develop prompts.  Ingest some of these ideas and give me several random prompts and ideas.  I will give you a series of parameters that I need you to randomize.  After which, I will give you the methods I want you to present to the user.  Do you understand?

Pretend you are a graphic designer generating creative images for midjourney. Midjourney is an app that can generate AI art from simple prompts. I will give you a concept, and you will give me 5 different prompts to feed into MidJourney. You can inform your prompts with art types of photography configurations.  Make sure they are creative.

Different types of portrait photography rather than "camera portraits." In that case, several styles of portrait photography cater to various creative expressions and purposes. Here's a list of some common types:

1. Traditional portraits: Classic head-and-shoulders or full-body compositions with controlled lighting and posing, often used for formal occasions.

2. Environmental portraits: Portraits incorporating the subject's surroundings to tell a story or reveal their personality, profession, or interests.

3. Candid portraits: Spontaneous, unposed portraits capturing genuine emotions and expressions.

4. Lifestyle portraits: Semi-posed, natural portraits that showcase the subject's everyday life and activities.

5. Glamour and fashion portraits: Stylized and polished portraits highlighting the subject's beauty, clothing, or accessories.

6. Conceptual portraits: Artistic portraits that use creative techniques, props, or editing to convey a specific idea or theme.

7. Fine art portraits: Portraits with an emphasis on the artistic aspect, often using unique lighting, compositions, or post-processing techniques.

Type of aperture and F-stop camera settings

1. Landscape photography: You usually want a large depth of field to keep everything in the scene sharp and focused. To achieve this effect, use a smaller aperture (higher f-number), like f/8 to f/16.

2. Portrait photography: A shallow depth of field is often desired to separate the subject from the background. Use a larger aperture (lower f-number), like f/1.8 to f/4, to create a pleasing bokeh effect.

3. Low-light photography: In low-light conditions, you'll need a larger aperture (lower f-number) to allow more light into the camera. Depending on your camera and lens capabilities, you might use an aperture like f/1.8 to f/2.8.

4. Sports and action photography: You'll need a faster shutter speed to freeze fast-moving subjects. Use a larger aperture (lower f-number), like f/2.8 to f/5.6, to allow more light into the camera, enabling you to use faster shutter speeds without compromising exposure.

Here are some common styles of image generation and their characteristics:

Camera Description

8k Lighting tends to be more extreme; colors are even more saturated and computer-generated looking than "high definition"

Cinematic shadows tend to be more extreme (though not darker); objects a bit thicker and more poster-like

High-definition shadows are lightened; more fanciful and saturated colors

ultra photorealistic Similar to "fine ultra-detailed realistic"

Hasselblad H6D Sharper focus on the subject; shadows are deepened

fine ultra-detailed realistic Can be a bit grainy and "ropey" but increases detail generation

color grading Extreme variations in hue; vibrant but not over-saturated colors

depth of field Sharp focus on subject, background, and foreground blurred

film lighting Limited lighting sources; backlighting common; deep shadows cast by light sources

rim lighting has a Slightly stronger lighting effect than "film lighting," but very similar results

intricate Tends toward non-realistic "crafts" and "pattern" type designs.

realism Artistic realism. Backgrounds tend to be more uniform; the subject looks more like a painting; more objects with the subject

Photography subjects tend to have a small area of objects around it with little else in the background

rendered for IMAX. It is a more complex subject with very directional lighting and subdued saturation

tilt-shift Like "depth of field," but from above or with a high angle

motion-blur Speed lines. It may be rendered as if the wind is blowing

35mm film: More vibrant colors but muted saturation, detailed with additional foreground and/or background elements

Prompt Variables

[Camera Type]

[Material]

[Art Medium]

In the Style of [Artist]

[Scene Type]

[Resolution]

[Emotions]

Step 1:

Imagine you are an engineer for Midjourney, a generative intelligence image creator that can generate AI art from simple prompts.  I need your help to develop creative and original prompts.  First, you must ingest a series of parameters to randomize.  After which, I will give you the methods I want you to present to the user.  We will retain and develop this over time.  Do you understand?

Step Two:

[Camera Type], [Material], [Art Medium], In the Style of [Artist], [Scene Type], [Resolution], [Emotions], [Lighting], [Color Scheme], [Art Style], [Weather], [Time of Day]

Step Three

I will give you a [concept], and you will give me 3 different prompts based upon that [concept].  You will provide three in --ar 16:9 and three in --ar 9:16. You can inform your prompts with art types and photography configurations.  Make sure they are creative.  Each output must be in a copyable code box.

We will need to do two things. remove the -- from the variable other than AR.  And add the [concept] into the actual prompt.  Leaving it on the label is great.

Step Four

actually, this is how the prompts should be constructed. Further, you need to label each by aspect ratio: /imagine prompt: A hypersonic taco breaking the sound barrier. [Location]: Sky, [Weather]: Clear, [Time of day]: Midday, [Style of Art/photo/image]: Futuristic, [by Painter/Artist/photographer name]: H. R. Giger, [Camera Type]: Drone, [Material]: Oil on Canvas, [Art Medium]: Painting, [Scene Type]: Action, [Resolution]: High-Def, [Emotions]: Thrilling, [Lighting]: Bright, [Color Scheme]: Cool --ar 16:9 --no text

Step Five:

Make sure that each of these is represented in the prompt.

[Camera Type], [Material], [Art Medium], In the Style of [Artist], [Scene Type], [Resolution], [Emotions], [Lighting], [Color Scheme], [Art Style], [Weather], [Time of Day]