Generative AI

Impact on Higher Education

Ray Schroeder
UPCEA Sr Fellow

image thanks to Andras Vas

Please bookmark this site for later use:

https://sites.google.com/view/unhpresentai

Impact of Generative Artificial Intelligence on Higher Education

First: A word about presentation format. For the past fifteen years, I have been "power-point-less." Rather than using a static, aging format, I prefer to create Web pages for presentations to assure that they are easily accessible, updatable, and always available. I will not be following every link. The intent is that this will serve as a reference meta-site for you on the topic.

Please follow along on your own device, skip ahead, drill down with the links, or visit at a later date to delve more deeply that which is useful to you.

Let's ask ChatGPT: just what is generative AI? (ChatGPT accessed 5/22/2023)

ChatGPT

Generative AI refers to a category of artificial intelligence techniques and models that are designed to generate new content or data that resembles, and sometimes even surpasses, the patterns and characteristics of the training data it was exposed to. Generative AI models are capable of creating original and realistic outputs, such as images, text, music, or even videos.

Generative AI models typically rely on deep learning methods, such as generative adversarial networks (GANs), variational autoencoders (VAEs), or autoregressive models. These models learn from large datasets to capture the underlying patterns and relationships within the data, enabling them to generate new examples that exhibit similar characteristics.

GANs, for example, consist of two components: a generator and a discriminator. The generator generates synthetic samples, while the discriminator tries to distinguish between real and synthetic samples. Through an adversarial training process, both components improve their performance, resulting in the generator generating more convincing and realistic outputs.

Generative AI has found applications in various domains. It has been used to create realistic images, produce music, generate natural language text, create deepfake videos, enhance or modify existing content, and even aid in drug discovery and molecular design.

While generative AI has brought about exciting advancements and creative possibilities, it also raises ethical considerations, such as the potential for misuse, manipulation, or the creation of misleading content. Ensuring responsible use and deployment of generative AI is an ongoing challenge as the technology continues to evolve.

How Education Leaders Think Generative AI Will Impact Classroom Learning - Jeremy Cesarec, CB&A 5/9/2023

Ray Schroeder, Senior Fellow at the University Professional and Continuing Education Association (UPCEA) and Professor Emeritus at the University of Illinois Springfield (UIS): “It is critically important that we in higher education embrace and fully engage in generative AI. Clearly, our students are entering a workforce in which they will be expected to work alongside these technologies. It is incumbent upon us to prepare those learners for that workforce environment as soon, and as completely, as possible. The careers of our graduates and the successful functioning of enterprises that employ them are at stake.”

https://www.cblohm.com/blog/education-marketing-trends/experts-pov-on-generative-ai-in-the-classroom/

AI Is an Ally: Saving Teachers Time with ChatGPT - Katie Novak, THE Journal

The time is now to welcome AI into classrooms as an asset and an ally. There are so many things that AI can do to support teaching and learning, from analyzing data for schools and districts, to enabling students to research and understand information in different ways, and — arguably the most important thing — saving time and energy for teachers so that they can focus on the parts of teaching that are less tedious and most important. ChatGPT has really taken the spotlight over the last few months. While it’s not a perfect tool, it has surely kicked off a bit of an AI arms race, and there’s plenty of tasks that it can support teachers with for their day-to-day.

https://thejournal.com/articles/2023/05/11/ai-is-an-ally-saving-teachers-time-with-chatgpt.aspx

Now, the Genie Is Out of the Bottle - There is no going back!

Researchers all over the world are working on newer versions and enhancements. The target of some is AGI (Artificial General Intelligence) - the holy grail of AI.

The AGI concept is that it can learn to accomplish any intellectual task that human beings or other animals can perform.[1][2] Alternatively, AGI has been defined as an autonomous system that surpasses human capabilities in the majority of economically valuable tasks.[3] Creating AGI is a primary goal of some artificial intelligence research and companies such as OpenAI,[3] DeepMind,[4] and Anthropic. AGI is a common topic in science fiction and futures studies. https://en.wikipedia.org/wiki/Artificial_general_intelligence

There is an expectation that ChatGPT 5 will include some aspects of AGI. Already, Auto-GPT shows the ability to create new prompts without human input, as needed, to accomplish an outcome that it has been assigned.

OpenAI Launches $1M Grant Program for Democratic AI Rule Decision-Making Ideas - ERIC HAL SCHWARTZ, VoiceBot.ai (5/25/2023)

OpenAI has set up a $1 million grant program to look for democratic ways of making decisions about AI system rules. Ten awardees will receive $100,000 from OpenAI’s nonprofit arm, OpenAI Inc., to develop and test these concepts as part of the startup’s larger push for designing a regulatory infrastructure around generative AI and its many offshoots. The new project launches just after OpenAI CEO Sam Altman’s appearance at a U.S. Senate hearing on generative AI and as some tech firms express concern over recent European Union regulation proposals. Meanwhile, the G7 announced last week that it wants to create an international standard for generative AI, with the “Hiroshima AI Process” producing a report by the end of the year.

https://voicebot.ai/2023/05/25/openai-launches-1m-grant-program-for-democratic-ai-rule-decision-making-ideas/

Around the halls: What should the regulation of generative AI look like? - Nicol Turner Lee, et al; Brookings (6/3/2023)

But where Congress could garner a very quick win on the regulatory front is to require consumer disclosures when AI-generated content is in use and add labeling or some type of multi-stakeholder certification process to encourage improved transparency and accountability for existing and future use cases. Once again, the European Union is already leading the way on this. In its most recent AI Act, the EU requires that AI-generated content be disclosed to consumers to prevent copyright infringement, illegal content, and other malfeasance related to end-user lack of understanding about these systems.

https://www.brookings.edu/blog/techtank/2023/06/02/around-the-halls-what-should-the-regulation-of-generative-ai-look-like/

How to use generative AI creatively in Higher Education - Bert Verhoeven & Vishal Rana, LSE Blog (4/4/2023)

Generative AI presents clear implications for teaching and learning in higher education. Drawing on their experience as early adopters of ChatGPT and DALL.E2 for teaching and learning, Bert Verhoeven and Vishal Rana present four ways they can be used to promote creativity and engagement from students. The emergence of generative AI and the release of ChatGPT has split the realm of higher education. Generative AI in general is often viewed as a cheating tool and a threat to academic integrity. Universities and academics have also shown scepticism about the use of this technology, with some institutions banning it outright. However, whilst a degree of scepticism is warranted, it shouldn’t hold us back from exploring potential positive applications of AI, especially as employers are already beginning to require AI skills in entry level roles.

https://blogs.lse.ac.uk/impactofsocialsciences/2023/04/04/how-to-use-generative-ai-creatively-in-higher-education/

Let's Ask ChatGPT: What are some creative uses of AI chatbots in education? (ChatGPT accessed 5/23/2023)

ChatGPT

Chatbots have proven to be valuable tools in education, offering various creative uses that enhance the learning experience. Here are some examples:

Personalized Learning Support: A chatbot can act as a virtual tutor, providing personalized assistance to students. It can answer questions, offer explanations, and provide additional resources based on the individual student's needs and learning pace.
Language Learning: Chatbots can help students practice and improve their language skills. They can engage in conversations, correct grammar and pronunciation, and provide vocabulary suggestions, creating an immersive language learning experience.
Virtual Study Groups: Chatbots can facilitate virtual study groups, connecting students with similar interests or studying the same subjects. The chatbot can help schedule meetings, provide discussion prompts, and even moderate the conversation to ensure everyone gets a chance to participate.
Assessment and Feedback: Chatbots can be used to provide instant feedback on assignments, quizzes, or practice exercises. They can evaluate answers, offer explanations for incorrect responses, and suggest areas for improvement, enabling students to learn from their mistakes in real-time.
Virtual Museum Guide: For subjects like history or art, a chatbot can serve as a virtual museum guide, providing information about exhibits, artists, and historical events. Students can interact with the chatbot to deepen their understanding and explore related topics.
Career Guidance: Chatbots can assist students in career exploration and provide guidance on different career paths. They can offer information about required skills, educational pathways, and job prospects, helping students make informed decisions about their future.
Gamified Learning: By incorporating game-like elements, such as quizzes, challenges, and rewards, chatbots can make the learning process more engaging and enjoyable. They can provide a fun and interactive way for students to review concepts and reinforce their knowledge.
Virtual Writing Assistant: Chatbots equipped with natural language processing capabilities can assist students in improving their writing skills. They can offer grammar and style suggestions, provide examples, and guide students through the writing process.
Revision and Test Preparation: Chatbots can generate practice questions and quizzes to help students prepare for exams. They can provide explanations for correct answers and track the student's progress, identifying areas that require additional focus.
Emotional Support: In addition to academic support, chatbots can offer emotional support to students. They can provide a safe space for students to express their feelings, offer encouragement, and suggest coping strategies during stressful times.

These are just a few examples of how chatbots can be creatively utilized in education. The versatility and adaptability of chatbot technology make it a valuable asset in enhancing the learning experience and providing personalized support to students.

Some leading gen AI apps

ChatGPT https://chat.openai.com/
- How to: https://zapier.com/blog/how-to-use-chatgpt/
Bing AI Chat https://www.bing.com/new
- How to: https://www.microsoft.com/en-us/bing
Perplexity https://www.perplexity.ai/
- How to: https://www.pcguide.com/apps/perplexity-ai/
Pi chat https://heypi.com/talk
- How to: https://digitalpilgrim.io/it/ai/inflection-ai-chatbot-pi/
Google Bard https://bard.google.com/
- How to: https://www.digitaltrends.com/computing/how-to-use-google-bard/
ChatPDF https://www.chatpdf.com/ How to: https://plainenglish.io/blog/chatpdf-com-chat-with-any-pdf-using-the-new-chatgpt-api#ideal-use-cases-for-chatpdfcom

Success with using generative AI chatbots is all about writing effective prompts! Prompt engineers (those who can get the chatbots to produce their most impressive outputs) are earning top salaries today. It takes practice.

How to Write AI Prompts: The Key to Better Outputs from Generative AI - Laura Starita, the Content Strategist

Consider these elements of an effective prompt:

Describe your goal output. Tell the AI what output you’re hoping to see. If you want an email subject line and body copy, say that. If you want it to be less than 100 words, say that, too. A blog post of 1,000 words or less? Specify it.

Describe the audience and style. Tell the AI who you’re talking to and what tone you want to take with them. Are they nature enthusiasts? Practicing physicians? Newly certified accountants? Do you want conversational prose, technical writing, or sector-specific terminology? Provide those details.

Include keywords or phrases you want the AI to incorporate. This is clearly helpful for SEO purposes but also to communicate preferences about the terms you’d like it to use.

Ask for special formatting. The AI can format your output as paragraphs, a numbered list, a table with content you can input to Microsoft Excel or Google Sheets, and so on.

Provide source material or examples. A prompt that includes a content source can help the AI understand what facts or details you want it to use and what tone or key message you’d like it to focus on.

Revise. If you don’t get what you expect from the AI, try again. If it is totally wrong, try a new prompt. If you find some of what it produced useful, tell it what you liked and provide instructions for revising the rest of it.

https://contently.com/2023/04/13/how-to-write-ai-prompts-for-generative-ai/

Try Refining Your Prompts with This Tool

What a Prompt?

Welcome to “What A Prompt” – your go-to tool for generating creative and enhanced prompts to achieve the best ChatGPT results! Designed as a quick and easy-to-use prompt optimizer…just select your prompt enhancer method, type in your prompt and hit generate! I recommend you start with "Prompt Enhancer" in the drop-down menu.

https://freshly.ai/what-a-prompt/

So, let's try out some AI apps to see the range of approaches and responses they provide.

Here's one prompt that might stimulate a good comparison between a couple of apps:

I am presenting to a group of faculty members at the University of New Hampshire about generative AI in higher education - what key points should I be sure to include?

Here's a second prompt: I am preparing to teach a 16 week 3 credit hour class about generative AI to freshmen and sophomores at the University of New Hampshire. Generate a syllabus for the class, be sure to include learning outcomes, 15 weekly lecture/discussion topics, mid-term research assignment, and a final paper.

Here's a third prompt: What do you recommend regarding crafting a prompt for you - include suggestions on wording, length, detail and precision of request?

Identify a .pdf that is online or on your computer to test ChatPDF.... ask for a summary, strengths, weaknesses and how to improve the document. Save your lecture notes as .pdf and ask ChatPDF to create ten short-answer quiz questions on the notes.

Suggest an individual situation or emotion regarding work (as in frustration, promotion possibilities, burnout, friction at work) for HeyPi !

Survey: 30% of College Students Used ChatGPT for Schoolwork this Past Academic Year - Arrman Kyaw, Diverse Education (5/26/2023)

Students reported using the tool mostly for English (49%) followed by “hard” sciences like chemistry and biology (41%). And of students who did not already have a 4.0 GPA last semester, 12% of all student ChatGPT users reported that their GPA increased from the fall 2022 to the spring 2023, but the majority did not see improvement. Users said that the tool’s advantages included its ease of use, simplicity, ability to help in organizational skills, and its ability to collect specific information and save time in researching. However, they also listed disadvantages, such as overreliance, inaccuracy, and potential to be considered cheating.

https://www.diverseeducation.com/reports-data/article/15448462/survey-30-of-college-students-used-chatgpt-for-schoolwork-this-past-academic-year

There is concern that students may plagiarize by cutting and pasting chatbot responses to submit as assignments in college classes. Certainly, this has occured. I asked Bard for suggestions:

There are a number of things that university faculty members can do to keep students from submitting AI chatbot responses as their own in classroom assignments. These include:

Using plagiarism detection software. Plagiarism detection software can be used to scan student work for similarities to online sources, including AI chatbot responses.
Instructing students on the importance of academic integrity. Faculty members should make it clear to students that submitting AI chatbot responses as their own is a form of academic dishonesty.
Creating assignments that require students to use their own knowledge and skills. Assignments that require students to think critically and creatively are less likely to be plagiarized from AI chatbot responses.
Talking to students about the issue. Faculty members can talk to students about the potential risks of using AI chatbot responses in their assignments.

In addition to these measures, faculty members can also take steps to make it more difficult for students to access AI chatbot responses. This could include blocking certain websites or using a proxy server to hide the university's IP address.

It is important to note that there is no foolproof way to prevent students from submitting AI chatbot responses as their own. However, by taking the steps outlined above, faculty members can make it more difficult for students to do so and discourage them from engaging in this type of academic dishonesty.

Here are some additional tips for preventing students from submitting AI chatbot responses as their own:

Make sure that students understand the assignment requirements. Students should be clear on what is expected of them in terms of the content, format, and length of their work.
Provide students with feedback on their work throughout the semester. This will help them to identify any areas where they need to improve and make sure that they are not relying on AI chatbot responses.
Be vigilant in checking student work for plagiarism. Use plagiarism detection software and read student work carefully to identify any suspicious content.

By taking these steps, faculty members can help to ensure that their students are submitting their own original work.

Ray's Observation on Plagiarism Detection and Generative AI (SEE UPDATE IMMEDIATELY FOLLOWING)

Much has been written about detecting plagiarism using software. This is a very tricky business. Even Turnitin promisesd only 99% accuracy - and now they say it is higher, perhaps much higher! The mistake of charging an innocent student of cheating can have devastating consequences. I believe pedagogy may provide the best approach.

Authentic assessment is one of the best strategies. In this case, you might have students conduct an activity and then ask them to report and answer questions about the experiences and results of that activity. This has the added benefit of providing experiences that may be directly applicable to the workplace and could be cited in job applications.

For example, you might just ask students to discuss details of their work to see if they really know the material they submitted. You can do that in class in person or via a synchronous session. If you are fully asynchronous, you might schedule a phone call or one-on-one Zoom to go over what they have learned at midterms and the end of the class.

One of the best suggestions comes from Karen Kenney of the University of Exeter writing in Times Higher Education: " 5. My favourite – if you can’t beat them, join them. Have students ask the AI to write an essay and then critique it. With this solution, you are helping students to embrace this 21st-century tool, to recognise its benefits and limitations, and to demonstrate their own knowledge as well." https://www.timeshighereducation.com/campus/evaluation-chatgpt-action

Here are a couple more relevant articles:

https://www.washingtonpost.com/technology/2023/04/01/chatgpt-cheating-detection-turnitin/

https://metro.co.uk/2023/03/16/can-plagiarism-detectors-such-as-turnitin-detect-chatgpt-18455877/

Turnitin’s AI Detector: Higher-Than-Expected False Positives - Susan D’Agostino, Inside Higher Ed (6/1/2023)

Turnitin’s AI writing–detection tool has a higher false positive rate than the company originally asserted, according to Annie Chechitelli, the company’s chief product officer. When the product was released in April, Turnitin promoted it has having a less than 1 percent false positive rate. Now, the company has not disclosed the new document-level false positive rate.

“We remain steadfast in our strategy to be transparent with the education community about our findings and progress,” Chechitelli wrote. Turnitin attributes the discrepancy to the difference between testing the company did in the lab and experiences users had in the real world.

https://www.insidehighered.com/news/quick-takes/2023/06/01/turnitins-ai-detector-higher-expected-false-positives

On a Related Topic - Some chatbots automatically provide citations (e.g. Perplexity), some include citations if you ask in the prompt (e.g. Bard), others must be asked in the following prompt to add citations to the prior response (e.g. ChatGPT).

How to cite ChatGPT - Tim McAdoo, APA Style

Example:

When prompted with “Is the left brain right brain divide real or a metaphor?” the ChatGPT-generated text indicated that although the two brain hemispheres are somewhat specialized, “the notation that people can be characterized as ‘left-brained’ or ‘right-brained’ is considered to be an oversimplification and a popular myth” (OpenAI, 2023).

Reference

OpenAI. (2023). ChatGPT (Mar 14 version) [Large language model]. https://chat.openai.com/chat

https://apastyle.apa.org/blog/how-to-cite-chatgpt

AI experts challenge 'doomer' narrative, including 'extinction risk' claims - Sharon Goldman, Venture Beat (5/31/2023)

Top AI researchers are pushing back on the current 'doomer' narrative focused on existential future risk from runaway artificial general intelligence (AGI). These include yesterday's Statement on AI Risk, signed by hundreds of experts including the CEOs of OpenAI, DeepMind and Anthropic, which warned of a "risk of extinction" from advanced AI if its development is not properly managed. Many say this 'doomsday' take, with its focus on existential risk from AI, or x-risk, is happening to the detriment of a necessary focus on current, measurable AI risks - including bias, misinformation, high-risk applications and cybersecurity. The truth is, most AI researchers are not focused on or highly-concerned about x-risk, they emphasize.(follow the link for the full report)

https://venturebeat.com/ai/ai-experts-challenge-doomer-narrative-including-extinction-risk-claims/

What is Bing Image Creator and how to use it? - Calvin Wankhede, Android Authority

Thanks to Bing, you can now use one of the best image generation AI for free. A picture is worth a thousand words, but what if you could generate an image with just a single sentence? That’s now possible with AI image generators that let you conjure up pictures with a simple text prompt. Microsoft has become the latest company to enter the arena with its Bing Image Creator platform. So what can it do and should you use it over competing image generators like Midjourney? Let’s find out.

https://www.androidauthority.com/what-is-bing-image-creator-3318945/

Supercharge Your ChatGPT Prompts With Auto-GPT - David Nield, Wired (5/22/2023) --- The next step toward Artificial General Intelligence!

Auto-GPT is the latest evidence for this: It leverages the power of ChatGPT to create an autonomous AI assistant, capable of taking on tasks and projects on its own and working through multiple steps in a job without you having to prompt it every time. In other words, it does a lot of the hard work for you, without you having to come up with your own follow-up responses or ideas. Think about everything you can do with ChatGPT, then imagine rolling that into a system that can supply its own feedback and make its own choices. Take coding for example: ChatGPT can output blocks of code based on your specifications, but Auto-GPT can manage an entire software project for you.

https://www.wired.com/story/chatgpt-prompts-auto-gpt/

Fast Track to AGI: So, What’s the Big Deal? - Ray Schroeder, Inside Higher Ed (6/7/2023)

Artificial general intelligence (AGI) is a term that elicits fear in some experts who cite far-ranging human consequences, and eager anticipation among those who see the potential for human progress. We see glimmers of the AGI capabilities in autoGPT and agentGPT. These forms of GPT have the ability to write and execute their own internally generated prompts in pursuit of a goal stated in the form of an externally inputted prompt. Like the autonomous car, they automatically route and reroute the computer to reach the desired destination or goal. Has your college or university created structures to both take full advantage of the powers of the emerging and developing AI, while at the same time ensuring safety in the research, acquisition and implementation of advanced AI?

https://www.insidehighered.com/opinion/blogs/online-trending-now/2023/06/07/fast-track-agi-so-whats-big-deal

Ray's Gen AI Rule of Two: Always run your thread of prompts on two (or more) different gen AI apps. This way you are much more likely to catch "hallucinations" or other artifacts that do happen, albeit rarely, on today's apps.

Looking Ahead - Sam Altman's Plan for an OpenAI Academy (Perlexity 6/2/2023)

Prompt: tell me about sam altman's plan for openai academy

PERPLEXITY

Sam Altman is the CEO of OpenAI, a research organization focused on developing artificial intelligence (AI) in a safe and beneficial way. Altman has proposed the creation of an OpenAI Academy, which would be a free, online school for teaching AI to anyone who wants to learn1. Here are some details about Altman's plan for the OpenAI Academy:

The OpenAI Academy would be a free, online school for teaching AI to anyone who wants to learn1.
The Academy would be modeled after Khan Academy, which provides free online education in a variety of subjects1.
Altman believes that AI is going to be a transformative technology, and that it's important to make sure that everyone has access to the knowledge and tools they need to participate in this transformation1.
The Academy would be open to anyone, regardless of their background or prior experience with AI1.
Altman has not yet provided a timeline for when the OpenAI Academy might launch, but he has expressed his desire to make it a reality1.

Altman has also spoken about the need for regulation of AI, and has testified before Congress on the subject2. Despite his belief in the transformative potential of AI, Altman has also expressed concern about the risks associated with the technology, and has called for careful consideration of its impact on society3.

techcrunch https://techcrunch.com/2023/05/26/sam-altman-shares-his-optimistic-view-of-our-ai-future/

nytimes https://www.nytimes.com/2023/05/16/technology/openai-altman-artificial-intelligence-regulation.html

businessinsider https://www.businessinsider.com/sam-altman-chatgpt-openai-ceo-career-net-worth-ycombinator-prepper-2023-1