Student Interviews

[Music]

[Alex] Hello. Today we're gonna be talking a little bit about some of the practical use cases for artificial intelligence, a little bit about how we use it in our daily lives, either as students or for professional development. There's been a lot of recent developments in artificial intelligence and a lot of (the) different applications. Or platforms in which you can use it. So we're gonna talk a little bit about that today. We are all members of the Leadership Institute and all part of Miami University Honors College. My name is Alex Coolum. I'm a biomedical engineering major.

[Sarah]And I'm Sarah Freeman. I'm a mechanical and manufacturing engineering major.

[Jay]I'm Jay Vo. I'm a computer science major.

[Naomi]And I'm Naomi Maurer, A biomedical engineering major.

[Alex]All right, great. So then, Sarah, I believe you've got our first question.

[Sarah]Yeah. So I think Naomi's gonna comment on this, but how might generative AI enhance professional learning in the corporate environment?

[Naomi]Yeah. So I think AI can help enhance learning both with just like self and personalized learning as well as in company training for self learning. AI can generate personalized learning plans for people who want to learn a skill in a set amount of time. And it can use like data that it can learn about the individual and their learning patterns and preferences, and it can generate customized learning materials and recommendations for that individual. And it can also help keep professionals up to date on the news like innovations and discoveries in their field, which can just kind of help them in their own work and perhaps, get some innovation for themselves and then for training purposes. It can help get rid of the language barrier for multicultural companies and it can help generate on screen translations during live training sessions. And then, I think it could also help trainers give feedback out to employees like a lot faster than what is currently being used, and it can also help employees get a better understanding with the quick feedback. And just kind of help speed along the training process for new employees or just new training within a company. And then it can also help companies identify any learning or skill gaps. AI can help kind of see where those are and also can help the companies to close those gaps. And then lastly, I think it could help companies employees by encouraging them to ask questions. They can ask AI questions that they might be too hesitant to ask another person and then they can use their immediate AI generated answer to either just go on about their day or ask someone in the company a follow-up question based off of what they learned from AI's answer. So that's kind of what I think AI can do in a professional setting to enhance learning. But then what do you guys think about how it can help students and enhance their learning?

[Jay]Yeah. So I believe that generative AI is an excellent, excellent approach for students looking to learn any new skills. So there's a lot of ways that you can use generative AI to learn new skills. Like you can ask it to ask it to for a study plan either like one week or three months of that. It can generate you a very detailed study plan that you can follow. And then you can ask it for some answers or questions for simple explanation like I often do like explain like four (level), hard concepts like recursion and all of that. And besides that you can also use generative AI to ask for quizzes, exercises, and (what I do a lot is) project ideas to learn new skills, because AI really (can) generate some really cool and interesting projects ideas that I can follow through. And with that said, I also think that generative AI makes our learning easier and overall enhance our learning. For me personally, AI helps me gather information quicker so that I don't have to do a lot of research. So that's helped me to take on a more higher learning curve subject and ,yeah. So another thing is that generative AI can help me get through a lot of learning. Like for me as a computer science major, I have to do a lot of project configuration to make a tool work and generative AI helped me get through that as well as reading long documentations or materials. I can use AI to extract information from that or just give me a summary. Besides, this is not not only from a viewpoint, a lot of studies have been conducted to show that AI really has a profound impact on student learning. According to studies on the effect of generative AI on students problem solving skills at Bartin University in Turkey. Basically they would divide a group of students into two groups taking the same course. One group will use ChatGPT for further learning enhancement like asking questions out of class and the other group will not have access to Chat GPT. In the end, both group will take a pretest, like before the course will take a test and then after that it will take another test to measure, like the problem solving skills. And they found that using Chat GPT students in that group score increases by 15% while the second group only increased by 4%. So that's something that really shows that Chat GPT can really enhance students' learning because Chat GPT can create a personalized learning experience through conversations. They know what you're missing. And so on. So yeah, that's in my opinion, ChatGPT is a very good tool to enhance any student learning experience. What about you, Alex?

[Alex]Yeah, I'd certainly like to echo a lot of the things that you said. I use generative AI in a lot of my classes to help enhance my learning. I kind of treat often large language models, Chat GPT, usually my go to. I treat it like Google. It is a tool to enhance my learning. So one of the things you talked about. Asking it to explain a concept in like 5 varying levels of difficulty is super helpful because I can sort of, (pick) the very difficult concept and start and it can explain it to me as a large language model in very simple terms and get more and more complex. And so I kind of work my understanding up to a higher degree using that. So that's certainly something that I have done many times and it's been very helpful for me. But also, a very unique use case that I found of generative AI is that, Generative AI tools can be a reflection of our failure of imagination and in a way and it can kind of showcase that that's when the real learning starts, is when our imagination fails. So to kind of put this into perspective, so these large language models, they're all trained on pre-existing data and so that's going to, well, in some cases give it bias and other things and that's a topic for another time. But, it's only trained on what currently exists, it's not able to actually come up with anything new. And so there's a study that the Harvard Graduate School of Education did, actually, and what they did was they took this hard kind of philosophical question and they asked their law students about, OK, well, how would you go about this case? What would you do? And they asked. All the students came up with, like a general plan and an approach. And it was a hard philosophical question that didn't have an easy answer. But all of them came up with roughly the same thing. And then they asked a large language model to do the same thing, answer the question and kind of formulate a plan. And all of the students were a little disappointed to find out that all of their responses were almost exactly the same as Chat GPT, and it was nothing remarkable. It didn't have a very awesome or helpful solution. And so they, in this case, ChatGPT kind of showcased their lack of imagination, their inability to think outside the box and formulate new solutions. And so there's a quote from that study from Harvard, and it says that we believe that AI language models like ChatGPT can act as catalysts and settings where predictable responses have repeatedly failed us, like climate change, race relations, income inequality, and more. They could indeed increase our productivity, not by providing us with better answers, but by confronting us with unoriginal and average of everything on the Internet responses. So this way we can move into the realm of new alternatives that Chat GPT and other large language models cannot predict. So in a way, Chat GPT can act as unintentional satire, showing us how insufficient and bland our solutions can be. So in a way, it enhances your learning by showing you. Maybe what's already out there by showing you the average, so you can start to think outside the box. And so that's a way that I've been using it recently is to help with my creativity to see what's already out there and start to branch away from it and get outside of that. So, certainly a different perspective and a unique take on it, but one that I think is valuable. But yeah, great thoughts Jay, as well. And then, so Sarah, I know you've looked into this a little bit, but how could. So there's a lot of. Mixed emotions about generative AI being used and especially among instructors in the classroom and either students using it or teachers using it. So what are some ways that generative AI could be used in the classroom, either by instructors or students?

[Sarah]Yes, I'll speak on that a little bit before I dive in. Thank you, Alex. That was (a) very interesting story from Harvard. And I think as a mechanical engineer, I'm really excited to see how generative AI kind of facilitates design work as we get, you know, maybe the bad ideas out of the way quicker. But for classrooms, we talked a little bit already about identifying knowledge gaps really quickly. So generative AI offers really quick feedback. So this can help both instructors and students identify knowledge gaps quicker, so as students have homework or essays or something like that that they want to just introduce to a large language model and kind of figure out, you know, where their misconceptions are with the content. That can be done a lot quicker and on a more individualized basis. So that's going to save some instructor time. And then in addition to identifying knowledge gaps, it can personalize learning based on learning style. So instructors have limited resources, limited time and how they can present the content. But if you can utilize generative AI, you can approach the same content from different learning styles. So some might be visual learners, others like audio, some are more hands on, so if the instructor kind of, guides the content of the instruction, but then generally, it can be used to cater it to the way the student best learns it. That's another possible use. But like you said, Alex, there's kind of a lot to be done in terms of finding which tools are best from (like) security standpoints, efficiency standpoints. There's going to be a learning curve having instructors figure out how to implement this generative AI and training them on it. And so this is kind of a common thread, I think, in the generative AI conversation, but we want to try to find tools that are easy to introduce and that when we put time into learning how to best implement them, that's going to pay off in ways that you can see for both instructors and students. So that's going to be a challenge. There are a lot of, like ChatGPT is a good one, just kind of for general use. There are a few others that you can use for creating lessons,lesson plans, a little bit more on the instructor side of trying to save time, and then hopefully you can also have it help with (like) logistical things like reserving classrooms. Or maybe like organizing groups based on how people work together and things like that. That again would just result in time efficiencies and maybe some better dynamics within the classroom. Yeah, but I think there are some really exciting things to be done, both using it inside the classroom and outside the classroom. The big question is just going to be how do we get there and how do we use it in ways that are going to benefit both instructors and students without maybe taking too much sunk cost in the training time.

[Alex]Yeah, of course. Some wonderful insights. Thank you. Yeah. So great comments from everyone. Thank you so much. This has been a great discussion for any of you watching who are interested in some of the things that we've been referencing. We'll put a list of links and materials that we've used to kind of source this conversation in the description. Please feel free to check them out. There's some really interesting stuff going on with AI all over the place, all over the globe and in every area of industry. So wherever you're in, whatever you're interested in, it can help you and it can impact you. So thanks for tuning into this. We appreciate it.

[Music]

[Music]

[Patrick] How's it going, my name is Patrick Hanley I'm a mechanical engineering major at Miami University part of cohort 11 and today we have -

[Sarah] Yeah I'm Sarah I'm also a mechanical engineering major from Miami, part of cohort 11, and we're gonna be talking about achieving individual and organizational value with AI, which is an article from the MIT Salone Management Review, and the goal here is to help us learn more about embracing AI in the corporate world.

[Patrick] Yeah, and with this I have some questions to ask you. The first one is how can AI act as a co-worker rather than a replacement?

[Sarah] Yeah, so one of the concerns about AI in the workplace is that it's going to take jobs and replace knowledge workers, but actually what it can do is be used as a tool. So, one of the ways it can do that is get rid of maybe more tedious tasks like scheduling um or planning like sending emails things like that, and allow for more time um being creative it can also help with more efficient training so it just help you be better at your job um rather than replacing the job, and then the article we're talking about actually found that most individuals see a AI as a coworker a job threat. It was about 60% um that saw that AI tools could just enhance what they're doing or uh increase like the quality of their work rather than replace what they're doing since you still have like that human creativity aspect, yeah.

[Patrick] Solid. Next how can AI increase individual value?

[Sarah] So, the article talked about three main areas where it would be increasing individual value. So the first was competency, so a little bit like I said before you can get personal personalized feedback on your work. Nationwide is currently using it um when they train their callers. They're using AI to give the trainees like really quick feedback and really accurate accurate feedback to help them be more competent in their job after training, and then autonomy is another area in which you can add individual value. So the individual worker can now achieve more, have access to more information, and you can learn more about past actions. So, with more knowledge and with um more information available you can get your own work done a little bit better, and then it can also help strengthen relationships. So, it can help with co-worker communication or communicating with other business partners and customers, which also helps add to individual value. So, a lot of people said that AI improved interactions within their team. Members said 56%, and then managers said that AI improved uh about 47% of teamwork, and then overall like across departments 52% of people agreed that AI can help Collaboration.

[Patrick] Next up, how does a company go about supporting AI’s addition to individual values?

[Sarah] So, a lot of this lies with the duties of a manager when you're trying to embrace AI in a team, so hesitancy is really natural when you're trying to adopt AI into a company, and that's something you just have to embrace and accept, so rather than trying to push people to trust it, you should just encourage people to use it, and then as they find how useful it can be and how it can help their day-to-day work, then that trust can kind of come secondary. So, Confidence from the manager that it is something that's trustworthy, but also just encouraging people to use it more so they can see what a great tool it can be. That's how you kind of go about beginning to foster that individual value with people in the company.

[Patrick] How does individual value contribute to organizational value, and how do organizational leaders achieve this?

[Sarah] So AI can provide both individual and organizational value kind of separate from each other, but ideally it is um helping those two things mix, so it's contributing to individual value that then serves the organization or organizational value that serves the individuals. So, organizations with employees who get value from AI are 5.9 times more likely to get financial benefits, so there's a lot of incentive there to try to implement it at the individual level, because then that will result in organizational value, and then like I said before doing at the individual level is going to involve uh encouraging employees that they can trust AI and showing them the tools that are available. So, AI has a lot of variety in its uses so being able to just implement it in everyday use for individuals and make them comfortable with it will then lead into more organizational value both financially and just using a good tool across the organization.

[Patrick] The last question I have here is what should organizations look out for when trying to utilize AI?

[Sarah] So, the the first one article mentions is avoid the lore of like shiny AI that's low value, so don't Implement AI where you don't necessarily need it, especially if you're putting the resources in to try to come up with like a safe implementation of AI that people can trust and trying to encourage people to use it. You might be wasting resources if you are implementing AI that isn't actually valuable, so make sure it's um like a quality system, it's a secure system, and that the value they're getting out of it is making them want to trust it and want to use it more. And then also avoid burdening workers to serve the machine, so don't Implement AI for the sake of AI and then make people conform to you know now we're using AI and this is what we're going to do, but rather, again, show them AI is the tool and then make sure that the leadership is comfortable using it and has plans on how to teach individuals how to use it so that it's implemented at that individual level with like as little time taken out of the day as possible, because people aren't going to want to adapt it if it's just like another chore in their day okay. And then lastly just make sure that the benefits offset the efforts at the individual level. So like I said if they're taking a lot of time to train and learn AI, make sure that ultimately them using it as a tool is going to save time in their day or produce better quality work or make things more efficient. Just don't do it for the sake of doing it.

[Patrick] Those are all great insights and definitely beneficial to helping people understand and utilize AI in the workplace. Thank you!

[Sarah] Thanks Patrick!

[Music]

[Music]

[Sarah] Hello everyone my name is Sarah Freeman, I'm a junior mechanical and manufacturing engineering major at Miami University, and I'm part of cohort 11 with the Lilly Leadership Institute

[Patrick] Hello my name is Patrick Hanley. I'm also part of cohort 11 of the Lilly Leadership Institute and am a mechanical engineer major here at Miami University. Sarah's got a few questions to ask me regarding some key findings from three Goldman Sachs articles, to support us in learning about embracing AI in the corporate world, which is a big project of ours this year

[Sarah] Yeah so our first question is going to be how can AI impact GDP?

[Patrick] Yeah so going off of this the article that applies well is “Generative AI could raise GDP by 7%”. So it kind of gives a little bit into it just in the title, but breakthroughs in generative artificial intelligence have the potential to bring about sweeping changes in the global economy, according to the Goldman Sachs research. As tools using advances in natural language processing work their way into businesses and society they could drive a 7% - which is almost 7 trillion - increase in the global GDP and lift productivity growth by 1.5 percentage points over a 10-year period, and this is due due to most workers employed in occupations have uh partial exposure to AI automation, and following the AI adoption, the AI can take tasks that they do in their normal work day and simplify them for them, and then they can use the freed up time towards other productive activities.

[Sarah] All right, and then what is different about the generative AI technology that we’re talking so much about now compared to previous developments in AI?

[Patrick] Yeah, so in the same article they talked about kind of three different levels of development, so the first level is software 1.0. This is where the code was written by humans - function by function - to perform one task at a time, and then it developed into software 2.0, which became the term for machine learning driven software development, where the main work was no longer actually writing the software, but it was collecting training data to train a network for a specific task. A lot of this was shown in automation of vehicles and how they can do like lane assist and all sorts of things along those lines. This is very labor intensive and expensive, so now we're here at software 3.0 which just came out. These have many capabilities right out of the box. The base models have natural language capabilities, reasoning, general knowledge of the world, but with this companies don't need to collect nearly as much data and it suddenly makes the technology much more useful, accessible, and less expensive as anyone can use it and it has so much access to all the knowledge out there.

[Sarah] And then we talked a little bit about how it'll impact GDP, but how is the labor market going to be influenced by generative AI?

[Patrick] Yeah so in particular Goldman Sach assumes that around 7% of workers will be fully displaced, but that most will be able to secure new employment in only slightly less productive positions. Also, the partially exposed workers to AI will experience a boost in productivity, and widespread adoption of generative AI could raise overall labor productivity growth in the US by around 1.5 percentage points annually.

[Sarah] All right, and then switching focus to healthcare specifically, why might Healthcare be ready for the disruption that's going to be caused by generative AI?

[Patrick] Yeah so the healthcare industry produces and relies upon massive amounts of data from diverse sources that creates a rich environment for applying AI, as it utilizes all that data for formulating its responses. The need for these technologies is giving inefficiencies in the healthcare system as it is estimated it takes more than 8 years and two billion dollars to develop a drug. With AI it will significantly decrease the cost of development and along with that the time to get a product to market.

[Sarah] And then how can chat Bots like chat gpt also be used in the healthcare industry?

[Patrick] Yeah so for chat GPT they can perform administrative tasks such as scheduling appointments and drafting insurance approvals, aiding healthcare professionals. All sorts of things like that, which will help the physicians get knowledge out to their clients faster, and it will also help fact check them as well. It has also been suggested that chat GPT could aid in clinical decision making down the road, such as diagnostics. This will take some more time for it to build trustworthiness and validation but through more development it will get there.

[Sarah] Great thanks Patrick that was really good Insight both from the economic side of things and healthcare. Thanks for your time!

[Patrick] Yes, thank you!

[Music]

[Music]

[Zach] I am Zachary copenhaver and I'm a computer engineering major here

[Brody] My name is Brody MacDonald and I'm a robotics engineering major with a minor in computer science

[Zach] We are going to be talking a bit about Communications between humans and machines and how AI can understand humans. So the first step in developing a useful generative AI is utilizing natural language processing to allow plain language inputs. How can this be achieved?

[Brody] So we can start utilizing natural language processing through - there are, it's like a - it's technically called like the acronym is NLP which is natural language processing, but basically we can set set up a like a model for an AI that takes inputs. You input a bunch of data in natural language for it to process, and some of this data can be processed and some of this can be unprocessed and the goal is to teach the AI how to read in a natural language based on however much data you can give it, basically, and a lot of fine-tuning and like it's going to it's going to require a lot of data to properly understand complete natural language, but over time through inputs and prompts and other methods it will slowly get better and better at understanding these natural language prompts and eventually it will be able to read almost any natural language that it has been taught, to a reasonably well extent. It also can, based on the data inputted, it can also respond in natural language and depending on how thoroughly you train the AI it will sound near human. There's obviously going to be some problems with that and it will not be perfect every time but depending on how much data is inputted it can be pretty close.

[Zach] Is there a point at which we can stop aiding the AI in learning these languages or where it can start learning them on its own?

[Brody] So to an extent, we can leave, once an LLM develops enough of a database that it has a like a a baseline understanding it can sort through data based on previous inputs and stuff. But, to an extent is really the issue, because as you input data it is going to learn continuously but there are things like errors can happen and if we allow an AI to self-learn basically, and it finds or like there there's an error that occurs it will basically propagate through the the rest of the data that's sorted and that can cause a lot of issues in terms of processing like natural language, so like if one chunk of data that's being inputted in like into a database is interpreted wrong by the AI it will use that as a reference for the next obviously data sets that are entered and it can propagate through the rest of the data that's used.

[Zach] Yeah, so like one bad input can kind of ruin a whole lot of the data processing.

[Brody] Yeah, it causes or it can cause a lot of issues. I know a lot of, once large language models get large enough they can basically be reliable enough that you don't have to monitor it, as with things like chat GPT. The amount of databases that are filled with correctly sorted data is large enough that it can catch an error and notice it and it will correct it based on that.

[Zach] When AI is given a prompt, what is parsing, and why is it so crucial to getting a proper response from the AI?

[Brody] So, parsing is what the AI will do when it receives a prompt or a data, like a set of data, and basically what it does is it goes through, basically character by character, and it will categorize them based on - it starts off with it's technically a pre-parsing step and it will go through and separate words punctuation, like pauses in a sentence, that sort of thing, and then the actual parsing step is where it will go through and it will label each chunk or block of characters and it will label it as either like a noun a verb and through that it will read in the data that is being entered, and it can convert it into something that it can understand and effectively give a proper response based on that.

[Zach] Okay and what are some ways that natural language processing is being implemented in businesses right now?

[Brody] So, a lot of companies are using Ai and chatbots with natural language processing for online help assistance. Like, instead of having an actual person or a technician sit on the computer and wait for an input from someone, they'll use Ai and based on the data that the company has inputted into the LLM it can effectively be faster and more accurate in terms of responding to a customer. So, if a customer has a very specific question about a very specific product a company sells, an AI will take way less time to sort through all of the company's records and find, not with 100% accuracy, but it will find an answer for the person much more rapidly than say just a sales technician. They're also using it for things like translation, so if a company expands globally and they don't have necessarily a large branch in another country at that point they can use AI to translate customer concerns or advertisements. They can put them into an AI and it will translate into the natural language of the region that they're in.

[Music]

[Music]

[Brody] All right so today we're going to discuss the impact of artificial intelligence on topics such as change management.

[Zach] I'm Zachary Copenhaver, a computer engineering major here.

[Brody] And I am Brody MacDonald, a robotics engineering major with a minor in computer science. So let's get started. How can leaders be prepared for the change brought by AI?

[Zach] So, leaders need to ensure that they have a very good understanding of how AI can help users at all levels of the company, from employees to managers to executives. They need to be able to effectively communicate the benefits and reasons for adopting AI across all levels of the company, and make sure that everybody - it actually helps everybody throughout the company in tangible ways.

[Brody] Okay, what kinds of change management strategies should leaders Implement to use AI to the benefit of the company and the worker?

[Zach] So, leaders need to have a bold vision for their AI strategy, and they need to tie it to the core business strategy of the company. They can't really just adopt AI because it's a big trend these days. They need to have a solid idea in their mind on how to use AI and they need to communicate that vision to their employees. That's something I'm going to be saying a lot here: communication. The strategy needs to be coordinated across the whole company and all levels of the company need to be actively involved in the strategy so that everybody feels like they're actually making a difference in implementing AI.

[Brody] Okay, what factors would you consider important in establishing an effective change management strategy?

[Zach] Well, like I said, communication is extremely important. Communication and commitment are the keys. The entire organization needs a very clear and firm understanding of your vision. Executives need to adopt a long-term strategy and fully commit to it. A company that is not fully committed or is hesitant to adopt to a new long-term strategy should not invest or jump into AI, otherwise they could get caught up in a situation like the .com bubble of the 90s and early 2000s where a lot of companies and investors took risks because of the new technology and they didn't really have a lot of solid plans for it, and then, of course the bubble bursts and a lot of people lost a lot of money from that. AI won't simply add profit without human effort to implement and use it. It provides opportunities and business leaders need to understand how to take advantage of those opportunities.

[Brody] How can leaders get workers to trust the change caused by AI will be beneficial for them and the company?

[Zach] So, leaders need to communicate with their employees. That's the big thing. They need to communicate the benefits of the change and why they are doing it. The employees need to know that their leaders know what they are doing, that they're not just stumbling into all of this change right away. They need to make their intentions very clear, and the leaders themselves need to help with the process of implementing the AI.

[Brody] What should leaders try doing if their workers are resistant to that change?

[Zach] So, leaders should be able to recognize and acknowledge the opposition to change and understand why there is that opposition. They shouldn't dismiss the concerns of their workers. They need to carefully address these concerns and the risks associated with AI. They need to show the workers that the leaders understand their point of view, and that they're actually trying to find solutions to these problems, and that they have everyone's best interests in mind.

[Brody] Okay. How can leaders create work environments that are not only accepting but welcoming to the change caused by AI?

[Zach] So, employees need to be prepared and empowered to accept and carry out this kind of change across the whole company. The leaders need to allow workers to experiment with AI on their own. Learn it, work with it, by themselves and come up with their own ideas for how to use it, and they need to be empowered and feel like they can bring up these ideas with their management and actually affect change in the company. The employees need to trust that the leadership has their best interest in mind and actually care about them and their career. They need to be educated on AI so that they actually know how to use it and understand the risks and how to use it responsibly and properly. The goals of the organization need to be tangible and achievable, something that the workers can actually see themselves working towards, and leaders should be able to foster this kind of accepting culture across the entire company.

[Brody] How can implementing AI affect soft work functions and the need for humans to fulfill these work functions? How can leaders address fears surrounding this?

[Zach] So, many job functions require a lot of human experience, expertise, and decision making, and many job functions can be automated by machines, and a lot of people are afraid that generative AI is starting to affect these kinds of soft functions, that AI can make them irrelevant or take their job, and leaders need to show that this will not be the case. They need to train people in order to use AI as more of an assistant than as a full worker. As a Helper and not a complete replacement.

[Brody] How should leaders manage the changes in business operations to effectively integrate AI?

[Zach] Leaders need to carefully reimagine their company and how it would change with the adoption of AI. Changes will naturally happen because of AI and just the flow of time, but leaders need to actively think about what changes are needed in the company now, and how they should happen, that leaders shouldn't just wait for the change to happen. They need to play an active role in thinking about and making that change a reality.

[Music]

[Alex] Alright. Hello. Well, thank you for joining me today, Micah. I understand that you've been doing a lot of reading about change management, especially when it comes to AI. AI is the buzz. AI is the new thing. A lot of companies have been talking about wanting to implement it, but they don't quite know how. So I'm happy to be talking with you today about that. So my first question for you is what should be the first step in implementing AI?

[Micah] Yeah, thanks for having me on. I'm excited to share some of the information that I have for today. And for the first step in employing AI is to really just realize and understand your organization's needs and goals. So this kind of means asking yourselves questions like what are the biggest challenges that the organization is facing? How could AI help solve these types of challenges? What are desired outcomes for implementing AI and once you have a good understanding of your organization's needs and goals? You can start looking in and specifying the AI solutions that could be a good fit, so it's important to really just involve yourself into understanding a wide range of stakeholders, including employees, customers, and supplies.

[Alex] Yeah, certainly a lot of things to consider there. So let's say I've now decided that my company, we need AI, we want to implement it. What are some ways that you can start to introduce AI into your businesses?

[Micah] Yeah. So there are tons of ways where you can introduce AI into your business. One of the ways you can just get involved is by starting with a private project. This kind of you know will allow you to help test out the AI and a controlled environment and then on learn from your experiences before scaling it up to the rest of the organization. Afterwards you kinda have to focus on the specific problems you wouldn't wanna try and like to implement the AI across the whole organization. Rather you wanna focus more on the specific problems that AI can help you solve. Once you have that, you wanna get a buy-in from the leadership. In order to get the support of leadership. It's pretty important just to get the senior leadership before implementing AI, and this will generally just help you ensure that the AI is aligned with the organization's overall goals and objectives. Once you have leadership, you want to communicate with the employees. Just make sure that the employees are informed and just be prepared to address any concerns that you know your employees might have, but the benefits that AI could bring to your organization. You also want to then further invest in the training and development of your employees. AI will likely change the way that people work, so you have to make sure that it's important to invest in training good programs to help employees, learn how to work, and develop AI effectively.

[Alex] Yeah, certainly lots, lots to think about for sure. You mentioned a few things about the employees and their importance. And so often in change management I hear a term thrown around called human-centered design. So why should we use this human-centered design when we are implementing big changes within this aspect of change management?

[Micah] Human-centered design really just helps us to understand the needs and concerns of people who will be affected by the change. So this basically enables us to design change initiatives that are more likely to be successful and sustainable. Human-centered design helps us specifically build buy-in for change. So when people feel like they've been more involved in the design of change, they are usually more likely to support it and implement it effectively. It also helps us minimize the resistance to change as well. Just by understanding the concerns of people who are affected by the change, it just allows us to address them early on and reduce the likelihood of resistance. Finally, On a final note, it just helps us to create a positive change experience for everyone involved. When people feel like they've just been hurt out and like, understood and respected, they're more likely to embrace the change. Human-centered design can make it work.

[Alex] OK. Certainly, yeah, sounds pretty important. So continuing that line of thinking, what sort of resources should we make available to our employees to properly educate them on AI?

[Micah] For sure. So there are also a number of resources that you can provide to your employees and properly educate them on AI. These can include training courses. These courses can help teach employees about the basics of AI, including more advanced topics such as machine learning and deep learning. Books and articles, you know, there's so many. Information nowadays available on AI and these resources can provide employees with deeper understanding. Online tutorials. Just the online tutorials available are just for so many of them. Like these tutorials can teach employees how to use the AI tools and technologies within the digital space. There's also the consideration of mentorship. Mentorship can basically pair employees with mentors who have experience with AI, and these mentors can really just provide employees with guidance support as they learn more about AI. Finally, just the access to AI tools and technologies. If you buy the software, you can really provide employees with hands-on access to AI tools and technology. Which you know should give them hands-on experience with the AI and learning how to apply it to their work.

[Alex] Yeah, certainly. OK, lots of good resources there. Now the big elephant in the room, the big question that a lot of people are asking is AI going to take my job? How do you address concerns of any variety when they arise regarding AI and change management in general?

[Micah] Yeah. So the concerns are obviously a big part about the change management. You kind of just want to be transparent. Being transparent with your employees about how AI is going to be used is such an important point because you want to understand why it's going to be implemented. You want to address these concerns that employees have about, you know, the common typical things such as job security, privacy, and ethics. Want to listen to feedback? You want to listen to their feedback by creating a safe space to share their concerns about change management and then use it to inform your AI change management strategy in the future. Finally, you kind of just want to provide the training support. You know having this software isn't going to be usable unless you actually train the individuals to use the software effectively, so you want to make sure they have the skills and knowledge to work with AI. And finally, you know, you just wanna motivate your employees to continue working through this change. That motivation is such a big part by, you know, just celebrating successes of this change, highlighting the positive impact that AI is having on your organization, which can, you know, help build support for AI and reduce resistance to change. Sure.

[Alex] Yeah. OK, great. That's wonderful. Thank you very much, Micah. So. Change often faces resistance, especially when it comes from higher-ups or an institutional change within an organization. But I know that collaboration among workers can really help to mitigate resistance to change. So how can I get my employees to start talking to one another about AI and how they can use it and help each other with it?

[Micah] Yeah, for sure. You want to make it as open and engaging and willing as possible for your employees to accept AI. So part of that is just by creating these opportunities for the employees to interact with each other. Nowadays, you know we have online collaboration tools. So using that, you know coming up with certain events, formal meetings. To maybe get people involved with a fun way like how an example could be just. Taking lunch to learn about AI and you know, just talk about it and engage your employees through this process. And furthermore, you kind of just want to provide employees with training on how to communicate effectively about AI. So training should cover topics such as how to explain AI in a way that is easily understood and how to address common misconceptions about AI. Furthermore, you want to just encourage employees to be open about their perspectives and hear from their viewpoints on their knowledge and experiences with each other. This could be done through formal presentations, blog voice posts, or just even the day-to-day informal conversation between employees furthermore. Providing employees with a safe space to just talk about their concerns and feedback about AI, making sure that their concerns can always be heard by the upper level management. Including HR so that. Their feedback really has an impact for making a difference in the change management. Finally, you know, as I mentioned earlier, recognizing rewarding employees. Who takes initiative to try to embrace the change in AI. For individuals who are having conversations, who are setting up these lunches, who are setting up informal meetings like. These can be done through public recognition, financial wars, or other incentives that bring people from outside their comfort zone, and accepting and encouraging employees to have these conversations about AI.

[Alex] OK, fantastic. That's wonderful. I love that little bit at the end there about getting outside of your comfort zone. That's something I'm a big proponent of and I think is really important, especially when it comes to organizational change. Well, thank you, Micah. That's all the time that we've got here today. But thank you so much for your knowledge on these questions, these topics on AI and how it relates to change management.

[Micah] Awesome. Thanks, Alex. It's great talking and just having that interview.

[Music]

[Music]

[Drew] Hello my name is Drew Laikin and today I'm going to be talking with Naomi Maurer, who is a biomedical engineering major at Miami University and in cohort 11 about some of her findings from the McKinsey's Global Survey and Deloitte’s research on generative AI. So, to kind of start off, are companies currently implementing generative AI into their operations?

[Naomi] Yeah, definitely. So generative AI has already shown to have a presence in the workplace. Even with less than a year since these generative AI tools debuted McKinsey has found that one third of the companies they surveyed are using generative AI for at least one business function, and these companies are implementing generative AI where the most value is, so the use of it is often found in the fields of marketing and sales, product and service development, and service operations like customer care and back office support. Generative AI is also a topic being spoken about during a lot of board meetings. 40% of the respondents said that because of the advancements in generative AI their companies are planning to increase their investment in artificial intelligence.

[Drew] So how exactly are these companies going about implementing AI into their workplace?

[Naomi] So, one of the most important features of an AI ready culture is definitely trust. To get past AI related fear employees need to believe that their company has their best interest in mind and that the implementation of the technology is for their benefit. Change management incentives and training activities are all ways companies are currently trying to build trust within their companies and Deloitte found that 37% of companies are currently using these practices to help people implement this technology into their work. Another way companies are implementing generative AI is by creating new job roles and functions to maximize the use of these AI advancements. They found that 38% of organizations are doing this already.

[Drew] So, what are companies trying to achieve by implementing generative AI into their daily work activities?

[Naomi] Yeah so it is well known that using generative AI can improve efficiency and reduce operational costs, which is the goal most organizations have when implementing generative AI, however, Deloitte also notes that organizations that have already implemented these generative AI tools and are seeing the most value from it are actually focusing on using generative AI with a growth mindset. They are striving to use these tools to expand their company by creating new products, improving customer satisfaction, and entering new markets.

[Drew] So what's something that companies should be wary of when implementing artificial intelligence?

[Naomi] Yeah, so one of the main pitfalls Deloitte recommends avoiding is pushing IT or data science people to drive the organization's AI transformation. They believe that it is the senior business leaders who should be taking the lead in embracing AI by integrating into their organization strategy, and McKinsey also says that routine maintenance checks will be needed to ensure that only quality and ethical approaches are used in the implementation of these AI tools.

[Drew] So based on what you've learned from what you've been reading, what should we as students do to prepare for AI in the future?

[Naomi] I think one of the most important things we as students need to do is to view generative AI with an open mind. We need to be open to the changes and retraining that will be inevitable with these AI advancements. Another thing we can do to prepare for the future is to start learning how to use it now. We should learn how to form good prompts and learn how we can implement it in our daily lives so that way we know how to use it and what to be careful about when we enter the workforce.

[Drew] All right well thank you for speaking with me today!

[Music]

[Naomi] Hello! I'm Naomi Maurer and I'm from cohort 11 of the Lilly Leadership Institute. Today I'm talking to Drew Laikin who is a junior computer science student also from cohort 11 of the Institute. We will be discussing some of the topics brought up in a podcast from the McKinsey Global Institute called ‘Generative AI: How Will it Affect Future Jobs and Workflows?’. So the first question I have for you, Drew, is why is everyone talking about generative AI? Why is it such a big deal?

[Drew] Well COVID-19 had kind of shifted the labor market back in early 2020 and 2021. There's a lot of disruption and remote work became a lot more common, and so that was one of the major contributing factors to the early popularity of some of these AI tools. AI's been around for a while but in that time frame, with the disruption in the labor market, it started to become a lot more popular in dealing with some labor shortages. Looking back there's about 9 to 10% of jobs every decade that were net new OP occupations. So these are jobs that hadn't existed before, and so it was kind of an inevitable shift but COVID-19 was definitely contributing factor in it.

[Naomi] And what could people expect about the future of jobs as the use of AI continues to grow?

[Drew] Well the impact of generative AI could automate almost 10% of tasks in the US economy, and there would be like an estimated increase of about 700,000 net zero jobs. These are jobs that have a net zero carbon emission, so you know jobs in renewable energy and green resources, which is a huge step forward in environmental progress, and yeah I think it it'll be a very positive thing for the future, but it'll change a lot of the ways we work.

[Naomi] Okay, so going off that, how might AI affect disadvantaged groups within the workforce?

[Drew] Yes, so there can be some ups and downs when it comes to how AI might affect different groups within a workforce so for example women are 50% more likely to be in an occupation that might have more disruption because of AI, so in customer service or sales, and so we might want to see federal programs to kind of help manage this so that there's not certain groups that are disproportionately impacted by this AI transition, as well as people in lower wage jobs. There might be more automation in those kind of roles so we have to be careful about how we embrace AI so that it doesn't kind of push these people out of their jobs, they can adapt to new roles, and I think that's going to be important into the future of how we use AI.

[Naomi] Do you think AI can improve diversity within the workforce, and how do you think that can happen?

[Drew] Yeah, so when we're talking about this disruption in the workforce that AI is going to cause you know a lot of people are going to have different jobs than what they were looking at years ago, but I think this could end up being a positive thing, you know there have been 50% more occupational transitions in the last three years that have resulted in workers having disproportionately higher wage roles, and so this is pushing people who you know wouldn't have had those kind of roles into higher paying roles which is always a good thing. And so, I think with this shift up there's a lot of opportunity for people to kind of upskill themselves and progress into new spaces that they haven't been into before.

[Naomi] Okay, and so then the last question I have for you is what do you think we as students should do to prepare for the future with the increased AI technology?

[Drew] Yeah, so since it's going to change a lot of the ways we work, I think a big thing is going to be, you know, how do you be adaptable? How do you learn new skills kind of on the fly and stay up to date on technology? You know there might be particular skills like writing and like programming for you or software development that might have some degree of automation to them, but if you can stay on top of how to use AI to kind of augment that work that you do and if you're always learning new skills and staying on top of the latest technology trends, I think you can definitely stay on top and really be be a competitive worker for the future.

[Naomi] Okay thank you so much Drew!

[Drew] Thank you.

[Hoang] Hello so my name is Hoang Nguyen. I'm a biomedical engineering junior at Miami of Ohio University, and today we have Michelle Ebu coming to speak to us about AI in pharma. Can you introduce yourself?

[Michelle] Okay so my name is Michelle Ebu. I'm a junior biomedical engineering major at Miami University, and yeah we'll be talking about, we'll be speaking on the article ‘How Pharmaceutical Firms can Use Data and AI with Confidence’ by Aspen Tech.

[Hoang] All right, thank you very much. So I have a few questions to ask you on this. First of all the article mentioned that very few pharma companies see themselves as having a strong digital culture. Can you start by telling us what exactly is a strong digital culture in pharma organizations?

[Michelle] Okay, so yeah, in the article it talks about strong digital culture, and he mentions it in like four parts, so strong digital culture is basically: one - confidence in adopting new technologies such as artificial intelligence, two - the confidence in using these technologies, three - the ability to make decisions quickly and at skill, and four - the willingness to take risks. So the article talks about how pharma industries have been able to implement all of this, except the willingness to take risks. Pharma Industries are very, should I say, they are very skeptical about taking risks. They never want to take risks, even the digital culture leaders, and whenever I say the digital culture leaders I mean the companies who are confident about adopting these technologies. So yeah, that's basically what a strong digital culture is: the confidence in adopting new technologies, the confidence in using these technologies, and, you know, the ability to make decisions quickly and willingness to take risks.

[Hoang] Thank you, and how would you say the COVID-19 pandemic affected either negatively or positively the use of AI in pharma companies?

[Michelle] So the COVID-19 pandemic destroyed the drug manufacturing process. The speed of vaccine development has changed industry expectations, and putting new pressure on manufacturers to deliver faster, but the supply chain has not responded quickly enough to this demand, and not only for vaccines, but also drugs unrelated to COVID-19, so making sure organizations can predict product demand and adjust output accordingly is important to successful collaboration with supply chain partners, but many companies still coordinate this using basic technology which does not enable centralized scheduling. To meet new industry expectations, organizations are trying to implement technologies that facilitate predictive maintenance, replacing equipment parts according to a set schedule instead of when they need attention, inevitably leads to unnecessary downtime, loss production, and inefficiency. So, ultimately the pandemic has all shown that all systems that run more rudimentary IT such as spreadsheets or paper cannot carry the industry forward, so pharma companies expect an even greater disruption in drug manufacturing in the future, and this has increased the need for more advanced technologies such as AI and machine learning to ensure survival.

[Hoang] Thank you, and how is AI currently being used in pharma?

[Michelle] So AI right now is currently being used for big data in the drug manufacturing process. It's being used in real time quality assurance that's being able to produce actionable data on a batch mid production, instead of waiting all the way till the end and potentially wasting resources. So it's also being used in predictive maintenance like I mentioned earlier, meaning that manufacturers, they receive data from sensors when individual components of machines need attention, and they can replace or repair them promptly and instead of doing the routine scheduled maintenance that leads to a halt in production, causing a delay in the drug manufacturing process. So it's also used in supply chain network scheduling, which is essentially being able to keep track of the products they need for manufacturing that are coming as well as the various labs and manufacturers they came from.

[Hoang] And so AI is clearly the key to derive value from big data, and I know that 60% of digital culture leaders surveyed say that technology such as AI can reduce pressures in the company, allowing them to bring new drugs rapidly and securely to the market, so what do you think is stopping it's adoption?

[Michelle] So I would say that some of the significant challenges that companies are facing right now is the lack of leadership skills and support. Team leaders lack the skills to get their employees excited about the concept of working with artificial intelligence. I would also say that there is an absence of an overarching strategy, meaning that the employees don't know what to do with AI, they don't know what to expect from AI, they don't understand the objectives of implementing AI, and they don't know what the future holds for them as a result of AI being implemented into their workplace. The third I would say is the change averse culture. Nobody wants to change. Everyone wants to stick to the norm. I mentioned earlier that most of these pharma companies indicated that they were afraid of taking risks, and this fear of taking risks to avoid failure would not allow them to implement new technologies into their organizations. It's a huge challenge that pharma companies need to address. They need to address their approach to risk. The risk reward ratio often deters many pharmaceutical firms from adopting this leading edge technologies such as AI. But yeah, introducing AI doesn't necessarily mean that you have to change your existing processes, it just provides a more - it just necessarily provides a more efficient way for you to do the same processes, but not enough firms know that, so yeah many perceive that there is a greater risk in adoption of AI than there actually is.

[Hoang] Yes, so what would you say are the solutions to these challenges and how can pharma companies drive the digital transformation strategies forward?

[Michelle] Yeah I would say first of all you need to get leadership support. Without good leadership you will struggle to create the right culture and develop a digital transformation strategy. Leaders need to understand that the value of emerging technologies is not just for R&D but also for drug manufacturing. They will be able to lead the company successfully in this transformation process. And also, promoting digital literacy throughout your organization is also very important. Also developing an overarching strategy like I mentioned earlier. Think holistically about the value of AI. Think holistically about how AI could add to all parts of the drug manufacturing process and make sure that the organization is aware of these benefits. Also, finally, you need to seek advice from people that are already doing it, be willing to open up your company, be willing to seek this advice from people that already using AI in their own pharma industries.

[Hoang] And you said that you were a student at Miami University right now, so based on the learning from the articles, what will you take as a student when you enter the workforce?

[Michelle] Okay so going into the industry as a biomedical engineer there's a lot I could take from this article. The article highlights the transformative power of AI and predictive analytics in pharmaceuticals. I would recognize the potential of these technologies in BME for tasks like patient risk prediction, personal treatment planning, and optimizing healthcare processes. The dynamic nature of AI and data technologies implies the need for continuous learning. I would approach the BME field with an open mindset, with a growth mindset, acknowledging that healthcare is an evolving field, and staying up to date with these emerging technologies such as AI is very crucial for success. Yeah. AI is continuously progressing and it's here to stay, so it makes full it makes sense to fully embrace everything that it has to offer.

[Hoang] Thank you so much for answering my questions today!

[Music]

[Music]

[Michelle] Okay so hi, my name is Michelle Ebu, I'm a junior biomedical engineering major, studying at Miami University and today I am with…

[Hoang] My name is Hoang Nguyen. I'm also a biomedical engineering junior in Miami Ohio University, and I'm here to talk a bit about the challenges and something discretion that business leaders need to know before they start incorporating AI into their daily life.

[Michelle] Awesome yeah thank you for the introduction. So my first question for you today is how do you ensure data privacy and security in AI applications, and what steps have you taken to address ethical considerations in AI?

[Hoang] Yes. Okay so for the first part of the question, data privacy and security in AI applications is something that needs to be put into regulation and something that needs to have backing by the law to ensure that they could have the best usage while keeping the company safe. For example there has been a judge who - a federal judge - who issued a requirement for lawyers to certify that they do not use AI to draft their filings without a human to check for their accuracy, and for the client client lawyer security check, so I think that's the first step that all companies need to think about to help put this into a law. And regarding ethical considerations in AI I believe that this is something that each company has to train their worker into. What we see now in AI tools and workflows is the first generation. This is the same as when 20 or 25 years ago when computers started being integrated into the workforce, so I believe this is something that needs to be introduced into the workforce as slowly and carefully as possible.

[Michelle] Okay awesome, thank you so much for that answer. So my second question is how do you handle explainability and transparency in AI decision making processes, especially in regulated industries?

[Hoang] Yes so I believe this AI transparency has to require internal changes in the company itself, as well as external collaboration. As AI tools evolve and become more intuitive business leaders need to identify how their workforce and existing systems can be adapted to AI, and AI itself is moving at a much faster pace than we can comprehend, so it's important to have a collaboration with others to best integrate it into the workforce. On the regard of transparency I believe this is something that business leaders have to educate themselves in first. They have to educate the skills to know what AI works for, because currently while 94% of business leaders feel confident about integrating AI into the workflow, 98% of them acknowledge that they need to have a better understanding of AI and the long-term potential of it.

[Michelle] Okay, so how do you manage potential risks associated with AI such as bias, fairness, and accountability?

[Hoang] So, AI is large model learning, so depending on the data that is passed through, it could give out like a lot of wrong data training. For example there was an eating disorder hotline that has shut down its AI chatbot because they were giving out bad and harmful advice. Also, there are companies who have had experiences using AI in their hiring process that could turn out having racist and sexist questions or a different way of discriminating against a certain group, so although AI is a capable assistant, it is still an assistant - something that humans need to stay in charge of and humans need to control. That being said, I believe that human prevention is something that all AI systems need. What I suggest is for them to incorporate a system with human checks for both accuracy and fairness.

[Michelle] Okay so now that we've addressed the potential risks associated with AI, how would you say we could approach the continuous improvement and adaptation of AI models to changing business conditions and market dynamics?

[Hoang] Yes so I think the most important thing that a business leader needs to know is to identify the risk that they may face when they're using AI technology in their workforce. So, there are three risks that is the most potential and apparent. That is inefficient training, limited organizational skills, and lack of understandings amongst the leaders and the workers, right? And I think to do this we need to establish that AI is not going to be a replacement for workers but rather technical skills and a technical tool, and the ability to deploy AI in the long term will be a crucial tool that all workers need to have - all knowledge workers need to have. And, another thing I think that we should incorporate is that we still have a lot of untapped potential for AI and generative AI especially, so this is something that we need to look forward to in the long term.

[Michelle] Okay, I have one final question for you. So you're a student at Miami University and you're an engineer, so based on the learnings from this article what will you as a student take as you're entering the working world?

[Hoang] Yes, so I think the most important thing that I take out of the article is that AI is a great tool to have for the future, that it’s another bullet in your arsenal, something that you should master. This is not something that will replace your technical skills, you still need to have required skill for normal engineer, but with AI in your hand it can help you a lot with paperwork, with other R&D development, and that it’s something that you need to have if you want to enter the workforce in a competitive work space.

[Michelle] Okay. That is awesome, thank you so much Hoang for your time today.

[Hoang] Thank you.

[Music]

Video transcript coming soon!

Video transcript coming soon!

[Music]

[Jessica] Hi, my name is Jesse Gentles, and I am a third year software engineer major at Miami University of Ohio and I am also a member of cohort 11 of the Lilly Leadership Institute

[Jay] Hi, my name is Jay, I'm also a third year of science major at Miami university Oxford and I'm also a member of cohort 11 of Eli Lilly Leadership Institute

[Jessica]This is an interview for the 2024 Leadership Institute embracing AI conference. The goal of the conference and project is to explore how professionals can embrace AI to work at their full potential. In recent years, we have seen a growing interest and concerns surrounding the implementation of generative AI in the workplace. Today, we're here to explore future work with AI. We're gonna start by going over how generative AI will impact our future workplace. So the first question I have for you Jay is how do you believe generative AI such as Open AI's Chat GPT is expected to impact human intelligence and decision making in the workplace?

[Jay] Yeah, so according to the article Gen AI and future of work by Cognizant which I will draw from a lot today. Generative AI can generate more data insights because it has the ability to compile a lot of data from multiple sources into one. Another thing it can do is it can add another depth of comprehension because it can act as another viewpoint so that you can compare your ideas with. And finally is that generative AI can generate consistent data which are very important for human intelligence and improving decision making.

[Jessica] Thank you! What do you believe are some ways organizations and businesses can start harnessing an AI potential as well as managing the risk associated with generative AI?

[Jay] Yea, so with the same article, they draw out some actionable steps for businesses to take. First, businesses need to start now, they need to start integrating and getting AI right into the business as soon as possible because integrating and getting potential, it needs time and they need to start early so that they can get through the (early) phase as soon as possible. Second is to start small where businesses need to make changes only to the small part where the risk is lower compared to the big part of the organization so it makes managing risk easier. Thirdly is organizations need to have an understanding of the AI models and choose they are using so with more knowledge there will be less risks. The article gave us many examples of models that businesses can use like the public model, expert model, and industry model that will each serve a different purpose so it's the business job to understand what model they are working with and use that.

[Jessica] What are some of the potential risks and challenges associated with the adoption of generative AI, and how can they be effectively managed?

[Jay] So, there are many risks associated with integrating generative AI. The same article talks about inaccuracies and biases, I can give an example about bias where the Amazon hiring algorithm has a high biased towards women because (of) the data it's being trained on only chose men for their job roles. Another thing is security where we have prompt injection where hackers can input malicious prompts to make AI generate sensitive business information. And in order to manage those risks. As I've mentioned, businesses and individuals need to have a clear idea of what the model is doing and what they are trying to solve using a model and in that way they can better manage the risk.

[Jessica] All right, thank you so how might generative AI impact the traditional departmental-based organizational structure that we know today?

[Jay] So, according to the Future Gen AI article, AI has a potential to transform the department-based organization of structure since a well-trained AI can take on multiple tasks with minimal training compared to humans who can only take one task and do well with it. So with that now businesses don't have to divide their teams into multiple departments. Instead we just have one team work together and let the AI do the other tasks so that we can focus on creativity and go to the common goal.

[Jessica] Right, could you let us know how generative AI will differ from traditional AI systems in its approach to task?

[Jay] So, traditional AI systems like detection and recommendation systems only react to input by following preset rules, while generative AI can create information even though it is not trained on those (data). For example, the article gives an example of traditional AI in a streaming service that can recommend a movie that you like but generative AI can, in seconds, write an original movie script tailored to your individual taste and request. So, that means that generative AI can handle a more complex task like code generating or customer service compared to a traditional AI.

[Jessica] All right, that is quite cool. What are some steps that you're taking to be ready for an AI driven future?

[Jay] Yeah, so some of the steps that I'm taking first is to keep up with AI through articles, research papers, social media like X and LinkedIn. Another thing is, I use AI a lot and I also integrate AI into parts of my life like use it for my personal coding project, use AI to study for exams, and use AI to craft some of my writings like email or job application. So by taking those steps, I am actively preparing and keeping an open mind for an AI driven future.

[Jessica] All right, and what is the main take away you hope users will get from this interview?

[Jay] The key takeaway from this is that generative AI has a potential to revolutionize how businesses operate. It brings many new opportunities but it also poses many challenges. So to integrate AI effectively businesses should embrace generative AI now so that they can understand the technology, manage the risks more, and be more prepared for a future where AI enhances human potential and collaboration.

[Jessica] Thank you for speaking with us Jay.

[Jay] All right, thank you!

[Music]

[Music]

[Jay] Hi everyone, my name is Jay I'm a third year computer and science major at Miami University of Ohio and I'm a member of cohort 11 of the Lilly Leadership Institute and today I'm here with Jessie.

[Jessica] My name is Jesse gentles I am a third-year software engineer and major at Miami University of Ohio and I am also a member of core 11 of the Lilly Leadership Institute

[Jay] So we are here today for an interview for the 2024 Lilly Leadership embracing AI project, the goal of this project is to explore how professionals can embrace AI toward work at their full potential. In recent months, we have seen growing interest and concern surrounding implementation of generative AI in the workplace. So, today we are here to explore those perspectives and strategies related to AI adoption. Let's first start off with the workforce attitude towards AI, so why do you think that workers are generally more receptive to changes that generative AI is expected to bring?

[Jessica] The Boston Consulting Group found that although Chad GPT, the poster child for generative AI applications only launched in November 2022, already 26% of workers say they use generative AI several times a week while 46% have experimented with it at least once. So many employees are getting exposed to this technology before they have to use it for work, therefore they are more familiar with the technology and more receptive to these changes.

[Jay] So could you elaborate on the significant difference in AI perception in terms of the concern and optimism between users and non-users?

[Jessica] Yeah according to CIO there's a study done by the Boston Consulting Group with over 12,000 Frontline employees, leaders and man managers worldwide and what they found is that the more workers used AI tools the less they were concerned and the more optimistic they were about the impact of these AI tools. Just 22% of regular AI users and 27% of rare users said that they were concerned, however 42% of non-users reported being concerned on the other hand 36% of non-users said they were optimistic about AI, compared with the 55% of rare users and 62% of regular users. That's a big gap so as you can see there is a significant difference between the responses from users and non-users of AI

[Jay] Yeah, so given that significant difference. What, in your view, are some of the most effective ways for companies to encourage the employees to accept the use of AI in their jobs?

[Jessica] The best way for companies to encourage their employees to accept the use of the of AI is to use organizational change management. According to Cap Gemini, multinational consulting company OCM is about encouraging and enabling people to use the tools in which their organization has invested so that they're able to achieve the true business value and return on that investment that the company has made. In this case companies would need to show employees what AI can do for them. This is important because it addresses this big question a lot of employees have about “what's in it for me”.

[Jay] Right, so on the topic of OCM what are some critical success factors that organization should keep in mind when implementing OCM strategies?

[Jessica] Some of the critical success factors that they should keep in mind is making sure that OCM is integrated with project management and not only is it integrated it has to be integrated in the planning or defining phases of a project rather than later. In the project, organizations can also go about this by creating a new but safe environment for employees to get used to these new tools and learn the different practices throughout the project. These leaders of the companies also need to make it their top priority to demonstrate their commitment to this transformation process. They also need to reward risk taken and incorporate the new behaviors into the day-to-day operations of the company and then one of the other key success factors to keep in mind when using OCM is that businesses should make the new solution desirable and relevant for the employees by making sure to present the big picture and outlining the goals of the company and illustrating how the solution will help the company to achieve these goals

[Jay] Right, and what are some steps that you personally are taking to be ready for an AI future?

[Jessia] Yeah so some of the steps I would say I take is just getting familiar with any AI tools I have my hands on. So that's using Bard, in my free time, ChatGPT. I like looking into how the different prompts change a response from, you know, an all right response to a really good one and then also keeping up with all the different articles that are coming out about changes in industry and about how people can go about embracing AI.

[Jay] Right, so before we end the interview I just want to ask what is the main takeaway that you want people watching to take out this interview?

[Jessica] I would say generative AI has the potential to change the workforce as we know it as companies move towards embracing AI in the workplace they need to focus on success will use in organizational change management for the smoothest transition.

[Jay] All right, thank you for your time Jessie and have a great rest of your day.

[Jessica] You too, thank you for having me.

[Music]

[Music]

[Ryan] My name is Ryan Holthouse I'm a junior computer science major and a member of cohort 11 of the Lilly Leadership Institute here at Miami. Today we're going to be talking about embracing AI and how we can all embrace AI in the workplace. I am here with Thatcher Lincheck, if you want to go ahead and introduce yourself, go ahead!

[Thatcher] Yeah, my name is Thatcher Lincheck, I'm a junior at Miami University studying mechanical engineering, and I’m also a member of cohort 11 of the Lilly Leadership Institute

[Ryan] Yeah well today we're going to be looking at an article by Accenture specifically entitled ‘A New Era of Generative AI for Everyone,’ so we'll just go ahead and jump right in. I’ve got a couple questions for you if that's all right. First things first is just that AI certainly is not a new concept, it's been around for quite a while in different forms, so I guess why do you think it's only now taking such a rise to global attention?

[Thatcher] Yeah so I think AI is taking rise into global attention now more than ever because of just the mass availability of these large language models, so things like chatGPT, Bard, and just so many other language models are all accessible by the general public. It's just a quick Google search away, bing search away, whatever you use to just find this language model, set up a quick profile, and then you have access to its whole model, one of the best models available on the market right now and you can just search whatever you want. You can talk to it, you can do data research, and just everything is all there and just accessible to everyone.

[Ryan] Absolutely, I think availability is a huge part of that. With generative AI and large language models there is kind of a debate between the ideas of consuming these models as they exist or customizing them to a company's direct needs so what pros and cons do you see or does the article kind of see with these two methods or approaches?

[Thatcher] Yeah so starting off consuming is this idea of just taking what exists and using it already, so we see some pros of that of just it's already built, it's easy to set up and just immediately use. You don't need to train it at all. Another would be that the output itself is high quality and already humanlike, so it's generated to fit like a human would be expected to write, so say you were trying to like write an essay or something you put in your prompt and you're like I want an essay on this. It generates that like you would say it, like a normal generic human response so that has it built in already so that's good. Some cons of course though of consuming is the ideas of you won't have the specific information companies will use, so like their private sources, their websites and collected data, so like company websites won't have the data of, say it's like a career AI that's helping you find your job, like a favorite job for you, it won't be able to find that from the big list because it doesn't have access to that website, it doesn't know what's on that. Then another one would just be language models are not representative of the brand when messaging customers, so there's not the brand identity with the response of the company, so it wouldn't understand like this is what the brand wants, not what a generic human would be. The response is too generic for what the company wants to use. So, then we have the customizing angle which is taking an AI or building your own and just fine-tuning that to fit your company's agenda. So, we have some pros for that of course is just it's more accurate, fine tuning the AI will allow for more accurate and relevant generation because it has access to your private data, it has access to the website and it knows what the company wants. There's reduced bias of it of course because it's being made by the company themselves so the bias is more what you would expect and want rather than someone else's bias being from a pre-trained language model. And then finally for some cons of course you have it's just a lot of money. It's a lot of time and money to train this large language model, so that's definitely something that companies want to think about before they invest into it. Finally just if there's proprietary data added, there's always a risk of a data leakage and privacy issue, and that's just a bad thing if all that data gets out because of course companies don't want their private data getting out.

[Ryan] Yeah. Well, Accenture, in addition to all that because I agree with a lot of that Accenture lists specifically banking, insurance, and software platforming applications as some of the largest or most promising open to automation application kind of business areas, so I guess, what about those tasks do you think lends itself so well to AI and what does that mean for worker productivity?

[Thatcher] Yeah so I mean I think for banking and insurance first, a lot of that is just going to be this idea of analyzing large amounts of data, which of course is AI’s strongest suit. That's what AI is for. It takes all that data and it combines it into one thing to just make it easier for you to read, it's that big jumble just combined together, so that's a big thing. AI can also be essential just as risk assessment, making sure that irregular patterns that could sense fraud or anything is found quicker, analyzing loan applications to make sure they're accurate, so avoiding like cyber security issues, attacks and all that with AI is very important with identification first and foremost. So then with software it's a lot of just, this AI can help you code, it can help you program and just understand what you're looking for from coding, you don't have to do all that research of like I don't know how to code this, how do I do that, you can ask the AI and it will help you, which is into this worker productivity idea is increasing the efficiency that workers have to get their tasks done faster and improve their decision making, where the AI tool can give them more time to think of ideas and just help them out there.

[Ryan] Absolutely, well I guess to kind of wrap us up for today, I would love to know how you're currently implementing AI into your life, how you're embracing it, and why you think it's important for all of us to continue to do so.

[Thatcher] Yeah of course. I mean, I think, so - yeah. So I'm using AI pretty much in my entire daily life. I'm using it to help me with some programming stuff of course, just some personal projects, some things I do just for fun, just asking it questions, always trying to just learn more about AI, and I think it's important for everyone to just do this so they get more familiar because AI is going to be this big thing that's going to just be everywhere. Everyone's going to be using it and that's so important to get started now and get used to what AI is doing so you can have this proprietary knowledge for yourself and just know what you're doing with AI and just be what companies are looking for, because this is going to be the biggest thing out there.

[Ryan] Absolutely. It's certainly not going away anytime soon. Well thank you so much for talking today, had a great time and I look forward to talking with you more about this in the future.

[Thatcher] Yeah thank you.

[Music]

[Thatcher] Hello my name is Thatcher Lincheck, I'm a junior at Miami University studying mechanical engineering, and I'm a member of cohort 11 of the Lilly Leadership Institute. I'm joined today by Ryan Holthouse to talk about the KPMG survey, ‘Generative AI - From Buzz to Business,’ so how about you introduce yourself Ryan and then we'll get to questions.

[Ryan] Yeah, well my name is Ryan Holthouse. I'm a junior computer science major here at Miami, also a member of cohort 11 in the Lilly Leadership Institute and excited to talk about this.

[Thatcher] yeah so for my first question I just want to know, so the KPMG survey goes over a lot of different data, so let's start with: 77% of executives are viewing generative AI as the most impactful technology. What do you think is influencing this decision and why does generative AI rank so highly?

[Ryan] Yeah I think for a lot of companies it's a matter of efficiency, growing their market shares, and just generally gaining a competitive edge. AI obviously presents huge opportunities across the board. You kind of have to look at it beyond just, you know, we're not making existing work go faster alone, we're also boosting worker capabilities to do more and automate some repetitive tasks, allow for more, you know, and that's just with busy tasks alone. In the article KPMG mentions countless sectors with different applications that they can benefit from generative AI. I mean in marketing and sales we could see it being used for generated advertising content or consumer interactions, and in scientific communities like medical fields or life sciences it can be applied to test out new molecules, new chemical solutions, see how things kind of work together. Regardless of the actual area there's so much room for generative AI to push business forward, and I think it'll become very clear over time that embracing AI is really going to show the difference between those companies that do and those that don't and you know market shares, all of that.

[Thatcher] Yeah I agree I think that's definitely an important idea, and I think just working on that is going to be big. So then, what barriers of entry do you think exist for implementing generative AI, and what do you think executives and companies will Implement to remove these barriers?

[Ryan] Yeah there's definitely a lot of big concerns. One big one is cyber security, at least that I see. Specifically the article cites lack of talent, cost issues, and unclear applications. As for unclear applications of AI, in terms of kind of removing that barrier, I think that that's something you just kind of have to play around with, test out a little bit. Obviously you know you don't know the full extent of how you can use it before you try it out. As for lack of talent I think that can almost be wrapped up with the cost issues. Not only do we account for the potential corporate licensing that comes with using outside large language models - obviously a lot are free but sometimes you need to consider bringing in corporate instances for security, making sure that your data is protected within an in-house instance - but you also need to account for the cost of either bringing in experienced professionals who understand AI platforms, who can really apply those within the business, as well as training existing employees, bringing them up to speed on that. You know, naturally that's something that's would also come with kind of bringing it in and doing some test applications of AI, some test embracing kind of on a smaller scale, but you know that's all something that needs to be considered. I think the biggest way to remove those barriers is just, you know like I said, test things out on a smaller scale, kind of ease the transition. Obviously a lot of companies want to really push to get this done, but kind of implementing on a small scale and then bringing in more bigger applications I think is going to be the best way to combat a lot of that as we're seeing in the article and beyond.

[Thatcher] Yeah that's a really good point. So finally just on KPMG itself, they talk about regulation of AI and specifically more regulations like the US government's AI Bill of Rights. How do you think that will impact executives and their willingness to implement AI solutions?

[Ryan] Yeah, I think the US Bill of Rights as well as other acts that exist like the EU AI acts are both really important pieces of legislation. They've been in the news a lot and I think the biggest thing that steers a lot of executives away from implementing AI solutions is, as KPMG suggests, the monetary factor that comes from the potential breaking of those acts. I mean for the EU AI act it requires companies to properly evaluate system AI risks and companies found in violation of that can face fines of up to about 30 million Euros or 6% of their annual income, so it's no small number. I mean especially with acts in place in the EU like the GDPR, data is a highly sensitive topic, so it's not surprising to see companies kind of wary of you know not wanting to push it as much with their data, and you know be a little bit more safe than sorry on the AI implementation. I think that's another valid barrier that exists as well, especially when money is involved and there's potential fines for that. It's just going to be, you know, figuring out how to get through that.

[Thatcher] Yeah I agree I do think money is always a big reason for a lot of things. So, wrapping up here, I just want to know how you currently are using AI and implementing that into your daily life, and how you yourself are embracing AI and why you think that it's important for all of us to continue to embrace Ai and just continue to use it and get familiar with it.

[Ryan] Yeah well myself I am somebody who, I like to look over work a lot, especially when I'm writing things, so for me my biggest application of AI has been kind of using it to look over my work, do some extra proof readings, see what I'm potentially missing in my work, and I think that's another big application for real world applications, you know, work tasks as well, rather than having to run something through seven layers of approval that's something we could, you know, maybe run through an AI client and have them clear that out. As for the importance of AI I just think, you know, as we've gone over there's so many opportunities for it, there's so many ways it can be applied, and you know while there is risks associated with it that we have to manage, I think that if we can fight through those the frontier for AI is just limitless.

[Thatcher] Yeah thank you so much.

[Ryan] Yep, thank you so much.