AI: Productivity Beyond the Hype

Speaking of AI…?

Is artificial intelligence a humanity destroyer or just the latest shiny object coming out of Silicon Valley? AI software creators and industry executives would have us believe AI is the greatest revolution of our time. We better rethink everything we do, because it’s moving faster than we can comprehend, they say. What’s more, they warn that in a few years, we will have achieved Artificial General Intelligence—software smarter than humans in every dimension. Then what???

The hype is an engine to drive attention and sales for new, and presumably better, technology leveraging AI. That software you purchased last year? No good, it doesn’t have AI. You need to upgrade. As Devin Coldewey observed in TechCrunch back in 2017: “‘AI-powered’ is tech’s meaningless equivalent of ‘all natural.’” Even with the proliferation of generative AI like ChatGPT, Coldewey’s notion resonates in many ways today.????

Experts have raised important concerns about the use of AI, especially in an educational context. Machine learning systems incorporate and sometimes amplify the biases they extract from the data they feed upon. Facial recognition software may work better with some skin tones than others (not to mention the many privacy concerns related to this technology). Courtney Plotts, Ph.D. and Lorna Gonzalez, Ph.D. have done some exceptional thinking about addressing the risk of AI bias for the EDUCAUSE Review .??

No, I don’t believe AI is on the verge of overthrowing humanity. And yes, there are some important and legitimate concerns revolving around AI that you should carefully consider before rolling out any large-scale, AI-backed technological solution. But the current generation of free and low-cost Large Language Model (LLM) tools, such as OpenAI’s ChatGPT, Microsoft’s Copilot, or Google’s Gemini, do offer sometimes surprisingly useful resources for personal productivity.??

Fluency with these tools has the potential to be useful in accelerating a number of common tasks. Plus, getting to know the software will help you make sense of the hype, mitigate the risks, and identify ways of employing these services for the benefit of your students.??

Example Use Cases for Increasing Productivity??

Here are a few ways I’ve used AI to help with my day-to-day work. Examples here use ChatGPT 4o, but you can get similar results from other systems. An important thing to remember is that the more specific you can make your request, the better the output you’ll receive. One trick I’ve picked up from my colleague Jacqueline Pitter is to instruct the chatbot to “ask me questions until you have enough information to complete this task.” The chatbots will often ask good clarifying questions that help you achieve better results.??

Getting Started on Announcements and Newsletters?

For many of us, drafting a message to a department or our campus is not high on the list of things we enjoy doing, and getting started can be a real impediment. When you’re feeling stuck, let AI get the ball rolling.??

I submitted the following prompt: “Write a newsletter from the IT department at Springfield State University. Start with a one paragraph article about how work is proceeding well with our new HR system, describing some of the benefits. Second, include a section about how we are adding new security features to our multi-factor authentication system to improve security. Finally, add a reminder that all systems will be down on July 18 for important upgrades. Frame it in a positive, upbeat tone that expresses the ways IT helps the campus be successful. Include a brief summary of the topics at the beginning, and a thank you from the IT staff at the end.”?

What you’ll get will not be a perfect newsletter, but a good start. You will need to read it carefully for accuracy and intention, but an LLM can get you from an idea to a draft in a few seconds, and then you can take it from there. (A sample response from ChatGPT 4o is shown at the end of this post.)?

First Drafts for Position Descriptions?

Somewhere out there is someone who really enjoys creating position descriptions, but I don’t think I’ve met that person yet. Try something like this: “Write a position description for a mid-level Microsoft SQL DBA for a university IT department. Include the typical tasks the person performs, and the knowledge and skills required for the position.” Add any specific requirements or skills you want the candidate to possess, and you’ll get a great head start on an otherwise cumbersome position description draft. Finally, if it’s pretty much what you want but you forgot something, such as some evening and weekend work required, just ask the LLM to rewrite the document adding those additional requirements.?

Creating Lists??

If you’re trying to brainstorm a list of items based on a certain topic, LLMs can be great for (again) getting you started.??

For instance, I was leading a workshop on norms in the workplace, and I wanted a list of common norms for the group to consider. So, I asked, “What are 10 norms commonly considered important for an equitable and productive workplace?”?

That prompt gave me a very decent start on developing the list the I needed. These models provide a consensus of the tens of thousands of lists incorporated into its learning model to produce a sample list. Is it the right list? Is anything missing? That’s for you to decide, but it's certainly a good start.?

Concerns and Cautions?

A common theme you may notice in the examples above is that I’m using the LLM to give me a draft or to get me started. I never provide a prompt only to copy and paste the AI’s response. Think of it more as the turbocharger to get you moving quickly than as the car itself.??

The tendency for LLMs to “hallucinate” and generate incorrect results is well-documented. There’s no “common sense” in LLMs, no knowledge of the world that underlies the text that it processes and uses to produce its output. At least for now, that’s work for the human mind.??

As mentioned above, there’s always a risk of racial, gender, and other biases built into machines created by humans. You need to read the output with sensitivity to those issues and proceed accordingly.??

Finally, depending on how you use the results of your work, you may consider if it is appropriate or ethically essential to disclose your use of AI.?

Disclosure: None of the writing in this article, other than the example below, was produced by AI. I like to write.?However, DALL-E did create the banner graphic....

Acknowledgements: Thank you to my colleagues at Vantage Technology Consulting Group for some useful editing and suggestions. This work is my personal expression and does not represent any official position of Vantage.

Example Output: Newsletter Generated by ChatGPT 4o?

Springfield State University IT Department Newsletter?

Summary:???

This month's newsletter highlights the successful progress of our new HR system, exciting new security features for our multi-factor authentication, and a reminder about the upcoming system downtime for upgrades. We are dedicated to supporting the success of our campus community through continuous improvements and innovations.?

Successful Progress with New HR System???

We are excited to share that the implementation of our new HR system is proceeding smoothly. This modern system offers enhanced efficiency, user-friendly interfaces, and improved data management capabilities. Faculty and staff can look forward to a more intuitive experience with faster access to essential HR services, streamlined workflows, and better support for our growing community. This upgrade reflects our commitment to providing the best tools to help our campus thrive.?

Enhanced Security with New Multi-Factor Authentication Features???

Security is a top priority for us at Springfield State University. To further safeguard our community, we are adding new security features to our multi-factor authentication (MFA) system. These enhancements will provide additional layers of protection against unauthorized access and cyber threats. The improved MFA system will be more robust and user-friendly, ensuring that your data and privacy are well-protected while maintaining ease of access for authorized users.?

Important Upgrade Reminder: System Downtime on July 18???

Please be reminded that all systems will be down on July 18 for essential upgrades. These upgrades are crucial for enhancing the performance, security, and reliability of our IT infrastructure. We appreciate your understanding and patience as we work to bring you the best possible service. Detailed information about the downtime and affected services will be communicated closer to the date.?

---?

Thank you for your continued support and cooperation. The IT department is dedicated to ensuring that our technology helps you succeed in all your endeavors. Together, we are building a brighter future for Springfield State University.?

Best regards,?

The IT Staff???

Springfield State University?