What is Generative AI?

Aside from the robots that are inevitably going to take over the world, applications of AI can be many and diverse.  We’ll go into detail about the two big applications that will be affecting higher education in the coming days, graphical and language, and then touch lightly on some of the many other ways that AI can be applied.

Graphical

Currently, AI is being used to create artwork based on prompts provided by humans who are interfacing with the AI.  Some examples of this are Midjourney, which was used to create the images in this article, Stable Diffusion, and DALL-E. AI Image creation takes a huge amount of image data along with keywords and uses it to synthesize a new image.  There are some things it’s not very good at like fingers. It can never seem to figure out how many fingers humans are supposed to have.  It is also completely dependent on human input and all of its biases.

Language Learning Models

Language learning models that produce text very similar to what a human might create are also gaining popularity.  Much like with graphics the language learning models use huge data sets of text that have been created by humans in the past as a model for creating their own language.  It can have conversations, help with coding, answer questions, and even write essays, poems, and syllabi for courses.

Other Applications

• Recommendations – when Netflix tells you what shows you might like to watch next based on your previous viewing history
• Face Recognition – this can allow you to log into your phone or put bunny ears on your head in snap chat
• Virtual Assistants – Siri, Alexa, and Google Home are talking to us all the time and the more they converse with us, the more they are learning
• Targeted Advertising – AI takes all the information it has about you and then shows you the things that you are most likely to be interested in buying
• Spam Filtering – Every time you mark a message or report it as spam, the spam filters get a little better at recognizing the spam on their own
• Games – AI has been playing games for years, a machine first beat a grandmaster of chess in the 90s
• Autonomous Vehicles – AI is driving those Teslas that people keep falling asleep in
• Microsoft Office – office will soon have integrated AI that turns reports and spreadsheets into presentations and other documents
• The applications go on and on

Well, who better to tell us what ChatGPT is than ChatGPT itself?  This is how ChatGPT describes itself if you ask it what it is.

As an AI language model developed by OpenAI, ChatGPT is a cutting-edge technology that has the ability to understand and respond to human language. It uses a neural network architecture that has been trained on massive amounts of data to generate human-like responses to a wide range of queries and statements.

At its core, ChatGPT is an example of natural language processing (NLP), a field of artificial intelligence that focuses on enabling machines to understand and interpret human language. By using machine learning techniques, ChatGPT is able to analyze vast amounts of textual data and identify patterns in language usage, which it then uses to generate responses to a user’s input.

One of the key advantages of ChatGPT is its versatility. It can be used for a wide range of tasks, from answering questions and providing customer support to generating natural language text and even assisting with creative writing. Its flexibility makes it a valuable tool for a variety of industries, including education, healthcare, and finance.

Overall, ChatGPT represents a major breakthrough in the field of natural language processing, and it has the potential to transform the way we communicate with machines. As AI technology continues to evolve, it is likely that we will see even more sophisticated language models like ChatGPT emerge, opening up new possibilities for human-machine interaction and advancing the boundaries of artificial intelligence.

OpenAI. (2023, March 14). [ChatGPT response to prompt: What is ChatGPT?]. https://chat.openai.com/…

• First and foremost, it is not more than the sum of its part. That is to say, it is not creating original thought.  It is simply taking the thoughts and creations of humans and feeding them back to us in a way that feels very human.
• It is not always accurate. While you will often get the right answers from AI, it can sometimes also make up incorrect facts or give you responses that are completely unrelated to the question that you asked.
• It’s not going to put us all out of business. In fact, it might make our jobs easier. Have you ever been completely stuck about something that you have to write? AI can give you a jumping-off point for getting started.
• It isn’t human. It sounds human, but it’s not human. It will remind you of that repeatedly if you suggest that it is anything else.
• It isn’t neutral. People often want to believe that if they ask a machine for an answer, they will get a sterile answer with no human bias. However, these systems are built using data that we fed into them – data that was created by humans.  So, if that data is biased then the output of the AI will be as well.

TurnItIn

UARK Faculty have access to Turnitin software. This tool will help make AI-generated text visible to instructors. Please visit our Turnitin hub to learn more about how to enable this tool on a given assignment and FAQs.

What do you do when you suspect AI-Generated Text?

If your intuition is telling you something isn’t quite right with a student’s work, you can follow these steps below to address your suspicions.

Step 1: Score the submission (hopefully using the rubric).

Step 2: If applicable, review the Turnitin AI Writing Detection report to see if any text was flagged. Please see  Faculty Guide: Assessing Turnitin AI Writing Detection Reports for Academic Integrity written by the Executive Director of Academic Initiatives and Integrity at UARK. 

Step 3:  If Turnitin was not used, or the detection tool did not flag text, review the submitted work carefully for potential red flags often found within AI-generated text outlined by Dr. Bryson in the article linked above.

Step 4: If you are still unsure of what to do next, speak with the student and have an honest conversation with them about your suspicions. Give the student a chance to resubmit after clarifying expectations (if these weren’t clear from the start).

• Please visit the UARK Research Guide that provides citation guidance for artificial intelligence tools in APA, MLA, Chicago Style, and IEEE.
• There is also an academic integrity resource page that provides UARK faculty syllabus guidance for generative artificial intelligence tools. Citation guidance directives are provided within the different options.

Course Outline: Unit 1: Introduction to Biology

The scientific method and the nature of science
Properties of life
Levels of organization in living organisms
Biomes and ecosystems

Unit 2: Cellular Organization and Function

The structure and function of cells
Energy transformations in cells
Cellular respiration and photosynthesis

Unit 3: Genetics and Inheritance

DNA structure and replication
Gene expression and regulation
Mendelian genetics and inheritance
Genetic variation and mutation

Unit 4: Evolution and Natural Selection

The history of life on Earth
The mechanisms of evolution
Evidence for evolution
Adaptation and natural selection

Unit 5: Ecology and Ecosystems

Populations, communities, and ecosystems
Biogeochemical cycles
Human impact on ecosystems
Conservation biology

Unit 6: Diversity of Life

Classification and phylogeny
Prokaryotic and eukaryotic diversity
Plant and animal diversity
Fungi, protists, and viruses

Unit 7: Scientific Inquiry and Data Analysis

Experimental design and hypothesis testing
Data collection and analysis
Scientific communication and peer review

Assessment:

Exams and quizzes
Laboratory reports
Written assignments and essays
Participation in class discussions and activities
Final project or presentation

Textbook: Campbell Biology, 11th Edition, by Urry, Cain, Wasserman, Minorsky, and Reece. Pearson, 2018. Note: This syllabus is subject to change at the instructor’s discretion.

Prompt: “Write me an essay that describes the themes of the Lady of Shallot and include quotes.”

Themes of Love, Art, and Isolation in The Lady of Shallot by Alfred, Lord Tennyson The Lady of Shallot, a poem by Alfred, Lord Tennyson, explores several themes, including the conflict between art and life, the power of love, and the danger of isolation. One of the primary themes of the poem is the tension between art and life.

The Lady of Shallot is a weaver who lives in a tower and creates beautiful tapestries. She is forbidden from looking directly out of her window, and instead views the outside world through a mirror. This allows her to create art without being distracted by the messy realities of life. However, when she catches a glimpse of Sir Lancelot, she can no longer resist the temptation to leave her tower and experience life for herself. Tennyson suggests that there is a danger in focusing too much on art at the expense of real life, as the Lady of Shallot ultimately sacrifices her artistic pursuits for a chance at true love. As Tennyson writes, “Out flew the web and floated wide; / The mirror crack’d from side to side; / ‘The curse is come upon me,’ cried / The Lady of Shalott” (lines 192-195).

Another important theme of the poem is the power of love. When the Lady of Shallot sees Sir Lancelot in the mirror, she is immediately captivated by him. She risks everything to leave her tower and pursue him, even though she knows that doing so will result in her death. Tennyson suggests that love has the power to overcome even the strongest barriers, and that it can be worth sacrificing everything for. As Tennyson writes, “She left the web, she left the loom, / She made three paces thro’ the room, / She saw the water-lily bloom, / She saw the helmet and the plume, / She look’d down to Camelot. / Out flew the web and floated wide; / The mirror crack’d from side to side; / ‘The curse is come upon me,’ cried / The Lady of Shalott” (lines 184-191).

Finally, the poem also explores the danger of isolation. The Lady of Shallot lives alone in her tower, cut off from the rest of society. She creates art in isolation, and as a result, she is unable to fully experience the joys and sorrows of life. When she leaves her tower to pursue Sir Lancelot, she is finally able to connect with the world outside, but she does so at a great cost. Tennyson suggests that while isolation may be comfortable in the short term, it can ultimately be harmful to one’s mental and emotional wellbeing. As Tennyson writes, “And at the closing of the day / She loosed the chain, and down she lay; / The broad stream bore her far away, / The Lady of Shalott” (lines 238-241).

In conclusion, The Lady of Shallot is a complex poem that explores several themes, including the tension between art and life, the power of love, and the danger of isolation. Tennyson’s rich and evocative language helps to bring these themes to life, creating a timeless work that continues to resonate with readers today.

Prompt: “Write me a poem in the voice of Lord Alfred Tennyson about teatime on the Titanic.”

Oh, the splendor of the Titanic,
As she sailed upon the sea,
And the elegance of tea time,
On the deck so fine and free.

The teacups clinked so gently,
As they passed from hand to hand,
And the aroma of the tea leaves,
Filled the air so grand.

The passengers were merry,
As they sipped their tea with grace,
And the sound of gentle chatter,
Filled the ship’s vast space.

But little did they know,
Of the fate that lay ahead,
As the iceberg loomed before them,
And disaster quickly spread.

Oh, the tragedy of the Titanic,
And the loss of life so dear,
But in memory we still cherish,
Tea time on that fateful year.

For it was a moment of beauty,
Amidst the chaos and the strife,
And a reminder of the fleeting joys,
Of this precious thing called life.

Perhaps you don’t want students to easily use a generative AI tool (such as ChatGPT) for an assignment. Use the suggestions below to get started with alternate assignments. 

• Focus on analysis of sources and information. Currently, generative AI tools cannot provide in-depth analysis incorporating sources. Instead of having students write an essay, provide an already-written essay (that you’ve written or from another source), and ask students to critique or give meaningful feedback that uses course materials. 
• Require the use of course materials (vocabulary, concepts, lectures, discussions) in an assignment. In most cases, these materials cannot be accessed by AI. 
• Create assignments that are scaffolded, with small steps building toward a final larger product or submission.  

Prompt: How do I write an assignment that can’t be answered with ChatGPT?

Response:

One way to write an assignment that cannot be answered with a language model like ChatGPT is to focus on tasks that require physical manipulation or observation. For example, an assignment that asks students to conduct a science experiment, create a physical prototype, or perform a task that requires fine motor skills would be difficult for ChatGPT to complete. Additionally, you could ask open-ended questions that require critical thinking, analysis, and synthesis, rather than simply providing information.

OpenAI. (2023, March 14). [ChatGPT response to prompt: How do I write an assignment that can’t be answered with ChatGPT?]. https://chat.openai.com/…

Generative AI tools will most likely impact students in their future professions, so you may be curious about how you can use the tools (such as ChatGPT) in your course curriculum. These suggestions may help you create your own AI-based (or assisted) assignment. 

• Have students query information in ChatGPT and then identify any errors and write corrections to these errors. 
• Let ChatGPT make a suggestion for starting, as inspiration, and then continue writing from that starting prompt. 
• Ask generative AI to provide an outline for a project, writing assignment, coding exercise, deployment plan, or lesson plan.  Have students use the outline as a rough draft but modify it based on their specific use cases and present a finalized co-authored outline, which would eventually lead to following through with the steps. 
• Have fun with generative AI as an icebreaker or brain break. Are you stuck and don’t know what to do next? Ask ChatGPT! Do you need creative inspiration? Ask DALL-E for a piece of art!  

Have the students read a piece of literature and create an image based on that literature. It could be a character or a scene from the work.  They should also write a brief summary about why they included details that they did and be sure to have them include the prompt that they settled on as well. This piece was created based on the poem “Anabelle Lee” by Edgar Allen Poe.

• Sample Syllabus Statements from a number of other institutions related to generative AI (Please note that these are not official University of Arkansas policies.  The University has not vetted each of these policies, and the University is not recommending any one of them for specific use.)
• TIPS Lunch – Faculty Session on ChatGPT: Assessments, Assignments, and Academic Integrity – https://scholarworks.uark.edu/wctfscpub/55/
• Curated and frequently updated collection of resources by Curated content on ChatGPT by Instructional Designer, Dr. Heather M. Brown – Tidewater Community College – Center for Teaching Excellence – https://padlet.com/hbrown293/ai-articles-on-chatgpt-and-education-itonvxbr22rpayy3 
• Inside Higher Ed article on Designing Assignments in the ChatGPT Era – https://www.insidehighered.com/news/2023/01/31/chatgpt-sparks-debate-how-design-student-assignments-now 
• Matt Bower’s video presentation on what ChatGBT is and how it can impact education – https://www.youtube.com/watch?v=92y_oOXvj6c 
• Microsoft articles on how AI will be built into Microsoft Office applications – https://news.microsoft.com/reinventing-productivity/ and https://blogs.microsoft.com/blog/2023/03/16/introducing-microsoft-365-copilot-your-copilot-for-work/ 
• Resource from the University of Toronto on ChatGPT andgGenerative AI in the classroom – https://www.viceprovostundergrad.utoronto.ca/strategic-priorities/digital-learning/special-initiative-artificial-intelligence/