Skip to content

On-Demand: AI & Data Literacy

Mar 9, 2023

In this one-hour webinar, we will demonstrate the fundamental AI & data concepts necessary to recognize in order to stay competitive in a complex business environment and become data literate.


Adam Karasick:Hi, and welcome everyone. We're excited to have you here today. My name is Adam Karasick. I lead up the data and analytics practice within the digital transformation group here at EisnerAmper. I'm delighted to be joined by Mechie Nkengla. She's the chief data strategist at Data Products.

I'm going to walk you through a little bit of what we're going to be talking about today, and then I'm going to turn it over to Mechie.

We're going to take you through a little bit of the definition of what we're here to talk about today of AI and data literacy. We're then going to talk briefly about the impact that this has, not only on you as an individual, but on your organization. We'll walk through different competency levels that you can think about for you and your organization. We'll define persona archetypes. We'll look at some of the different types of people that are going to be involved in this process for you and your organization, and what they look like and things that you can do to start crafting that.

And then we'll look at a roadmap strategy for you moving forward. Where do you start taking your next steps? And finally, we want to leave you with some takeaways from this presentation today, what you can start thinking about as you move ahead, because it can be overwhelming. And we're going to try to make this as simple as we can for you. So I'll turn it over to Mechie.

Oh, you're on mute. Mechie, you're on mute.

Mechie Nkengla: Thanks, Adam. Thank you for the warm introduction. Welcome everyone. Very excited to be here and talk to you about AI and data literacy. So we start with our disclaimer. It is clear that the information that we're presenting is not legal or financial advice. These are views, opinions based on research and experience, and we're going to try to make this as exciting as possible for you.

And one thing, please feel free to interact. We love interactive webinars. I think they're a lot more engaging. So if you have a question ask, and we're going to try to see if we can answer as we move along. Move on to the next slide, please.

All right, so let's start with a polling question. What does data literacy mean to you? And remember, you can select your answer directly in the slide. You can just click on it, and it would register. So we're going to give you a couple of seconds here to see, to get a sense of what people think.

Adam Karasick:We're genuinely interested here just to see where everyone is starting from in this presentation. And then hopefully, as we move forward, we'll have a little bit of a different approach and take away from this presentation.

All right, Mechie, what do you think? Do you think we can take a look at that results?

Mechie Nkengla:Let's give a couple more seconds. I see 62 people have responded, but there are a few more. Let's give them a couple more seconds to see what that looks like.

Okay, I think we can go ahead. Adam, how do we get to the results of this?

Adam Karasick:We advance here. We can see.

Mechie Nkengla:Fantastic. So we see that is sort of almost not quite uniformly distributed amongst all the answers, but we're excited. This is, I guess, you thinking about what it means to you and that's critical in terms of mapping up a plan for yourself or your organization. Move on.

So what is AI literacy? AI literacy, it refers to the ability to understand and effectively communicate about AI technologies. Just as we learn to read and write and communicate with others. AI literacy is becoming increasingly important in technologies that are prevalent in our everyday lives. So by improving our AI literacy, we can make informed decisions about how we use this AI technologies and ensure that the way we use them is a benefit to society at large.

And that is critical, I want to point out here. So it's not just about understanding at a high level what AI is and understand how to leverage them within the tools and technologies that we have, but it's also critical to understand the impact that they have on us as a society or an individual.

So if you think about the common AI and machine learning technology, that are around natural language processing, computer vision, robotics, and you're leveraging to perhaps say a smart refrigerator, yes, it's nice. It makes your life a lot more convenient. But what are the impacts to your home? Is it listening? There are critical things that we have to be aware of as we think about AI technology, and that is part of the literacy as well. Move on to the next slide.

Now. Let's talk about data literacy. We just talked about AI literacy and the data literacy, we did a nice poll to set this off. Data literacy, again, is the ability to read, write, and communicate about data in context, very similar to what AI literacy is. And it is absolutely important that we... Data is really a part of every of our lives right now, from the way we... We live in the digital world. Every organization right now is taught to be a technology company that happens to do such and such. And the underbelly, what underscores digital is data. And so we're using it whether we think about using it or not.

And so it is critical that we can understand what this needs and why it matters. And taking a shift back to AI, we see that the AI market is projected to get up to about $407 billion in 2027. That's, what, four years away from now? $406 billion, let that sink in, and in 2022, last year, it was about $86 billion. That's a compound annual growth rate of about 36%. So it's going to be, it's huge. And so we're not asking everyone here to become a data scientist or a data engineer, but more of understanding at a high level the concepts and impacts. That's what's critical here. And of course it's going to have an impact in terms of how you work in your career as well.

Now let's move on and take a look about how the literacy actually matters in our everyday lives. So when we talk about data and AI literacy, there's usually this sphere. That's what I find anyway. So a lot of people have this innate sort of reserve about, "Oh, I'm just going to have to learn some statistics." No, the good part is, no, you don't really necessarily need to understand statistics, at a high level, at a deep level, anyways.

We use data already. It's ubiquitous in the way we interact with it. Unfortunately, we still have this reservation about what it means, but we use it every day. We are looking for a mortgage financing, everyone's researching to see what the best rates are. We want to send our kids to school. We're looking for what it means for that for school in terms of how far it is, what it's going to cost you, what the program is. Is it going to be best one for your child or your ward?

Your online shopping, everyone's cost comparing where you can get the best price. You're looking at your medical records, everything is digital, you can pull it. How many people here have fantasy baseball leagues? I'm sure quite a bit, right? And I've seen some of these that have really, really advanced models baked into it that does predictions and numbers and statistics about their leagues. And yet these are the same people that we walk in and have this huge reservation when anything comes around statistics or numbers.

But the first thing is to become comfortable with this. We already use this, and so we need to switch that fear or that reservation, because we're already using it when it's hidden underneath a user-based technology. What about our communities? So it's critical, when talking about AI and data literacy, it's going to affect our lives. And when we're talking about the lives, the way I like to think about it is it's going to affect us personally. It's going to affect our society, our communities, and it's going to affect the organizations, our workplace.

So we talked about the personal impacts. Now, let's get a little bit into the communities. So data impacts communities, there's no doubt. And so this is something that we should be conscious about highlighting and not hiding the experiences of the people in these communities, in order to ensure that there are equitable and fair outcomes from whatever decisions or policies from made out of this data. And I'll give you a good example.

Now there's this fantastic non-for-profit, that builds wells in developing countries. And they visited a village in some country in Africa and they said, "Oh my goodness, you guys, the women here are the ones that are responsible for bringing water. And so they have to make an hour's trek each way, every single day with big basins and buckets of water on their heads, and children on their backs to bring water back and forth. That's a lot of time. I bet if we built a well directly in the heart of the village, it will give these people a lot of time back and effort. And it's going to make their life significantly better."

And so they decided, "Yes, we're going to build the well." And they worked hard and got some funds and built the well. And then a year later they came back to take stock about how the lives of these villages have improved. And they started doing interviews with the women, because they're the ones that were responsible for fetching the water. And what they found out that the women did not like the well. And the question was, "Why? We thought this makes your life easy."

And then the response they got from there women villagers were, "Look, our lives are hard. We wake up from 5:00 AM from sun up to sundown. We're always busy taking care of people, the food, the children, the education working the farms and the gardens. The time that we took to go fetch water, yes, it was long, but it was our time that we were not responsible for someone else. We could mingle. We could talk. We could build a community amongst each other. And basically by you building this well in the center of the village, you've taken that away."

And this concept of where decisions are made from data without including the society or community that's going to be impacted by it, I call that data colonization. Another easy one to think about is in a scenario that's going to be more relatable at work. Say, you're line of business subject matter expert and you have a data science team or IT team comes to you and say, "We have built this fantastic solution that's going to optimize your work. It's going to speed this. You're not going to need to do this." And it's just plopped right there in front of you.

Now, this IT organization has got the data from some ERP or CRM system. They've understood from their perspective of what it means, and they basically served a solution to you without having you be part of that process, in terms of developing that solution. And there's always like, "Oh my goodness, you don't really get this. You're not the subject matter. You don't understand how we go about doing things, the reservations that we have." Or there are some issues that are not necessarily captured in the data but are more contextual that you're missing. So that's another example of data colonization. So we need to be careful about that in communities.

So when we talk about data in the workplace, it's important to understand the impact. We talked about AI and how it's going to be about $407 billion by 2027, compound annual growth rate. But in terms of the data literacy, this are some predictions I have here. It's going to be the most in demand skill by 2030. That's not far off. We're going to have future workers be lifelong learners. As new technologies come out, we're going to need to understand at a very high level, we don't need to be experts in the field themselves, but we need to be experts in using them, communicate with them, and understand them within contexts.

85% of executives believe that it would be as vital in the future as ability to use a computer is right now. There's no one that we can get a stable career work that cannot really interact with a digital computer in some fashion. And that's how data literacy is going to be.

So if we looked at the past, the way we thought of work, an organization was a factor of three pillars. We talk about people, processes, and technology. So we think about an organization, we think of the organization just made up of people, process, and technology. And one has been in business long, we know this for the last several decades, this has been the way we know things.

There's been a shift. As people become more digital transformed, things are more digitalized, we've seen that data has become an emerging, and is still emerging, key aspect. We still have the people, processes, and technology, however, this is all underscored by data, because data is sort of the glue that holds all these processes together. Adam.

Adam Karasick: All right, thank you, Mechie. So I want to talk you all through technology with all of the steps in the journey, as I'd like to call it, with regards to data and AI. I posed the question on the screen here, where are you in your journey? And you may be in different areas in this journey at one time, but when I first started going through this process myself, it was overwhelming. So I want to walk you through each of these steps as they're visualized on the screen, what they entail, how technology can help you through those processes. And then on the next slide, I'm going to talk a little bit about how you can take a leg up against the competition, against your peers by being very well aware of all the different uses of these with regards to technology.

So first, that lowest step on the left, data collection. I think if you start your journey asking questions about what you're trying to accomplish both individually and for your organization, that will better allow you to start collecting the right and the most appropriate data as you make your way through. And you have to remember, data collection is a lot easier today than it was, say, five or 10 years ago. Your mobile device, wearables, apps that are easy to develop will allow you to start collecting more and more data, because as many others have said before me, "Data is your best asset." And you want to make the best use of it, but you can't really go and analyze it if you don't have it. So collecting your data through these means is paramount towards moving through this process.

That second step, cleansing, that's where I tend to spend most of my time. Which I personally can't stand, but it's becoming a harder and harder thing to get through, which, again, allows you to make your way through this journey. But data cleansing often is the thing that holds up a lot of organizations. Messy data, incomplete data using tools that are now readily available, some of them in the AI space, allow you to better clean up, structure that data, fill in the gaps where your asset is really holding you back.

The third piece here is the part that I like the most is the analysis itself. With regardless to technology, there's a lot in the marketplace right now that allows you to better analyze your data and not be this advanced data scientist to be able to sift through it all. There are data visualization applications. There are programming languages out there. There's something as simple as the most efficient use of using Microsoft Excel. So there are a wide array of tools out there.

It's helpful to know what's out there, but you certainly don't need to be well immersed in each and every one of these. I know I was overwhelmed when I first started seeing the whole landscape of applications that are out there. But just knowing what's available to you, so that you can make the right conversations and questions, will allow you to start going through that asset that you have, your data.

The fourth step here is the collaboration piece. It's especially important now than it was ever before that you're connecting with your colleagues, your clients, and just really being able to share and receive information back and forth as quickly as you can. I think we're past the days now where you send an email and wait for response and you have to worry about time delays, version controls. There are tools and applications out there that allow you to just really connect right in with those that you want to talk to and work with.

Fifth, here's the stories and insights, which can be a fun piece. I talked briefly before about data visualization. This is the ability for you to really share and provide the output of your analysis to your end users, your stakeholders. Creating infographics, creating live dashboards, these are the things that really will set you apart from your peers.

And then finally, I can't leave out the notion of data security and how important that component is. You can go through this whole process and then something can go wrong, you might lose data, you might be breached. It is extremely important that that is always an important pillar in this journey for you.

So we talk about the competitive advantage with these six things. I'll briefly walk through some of them for you. Data collection, how quickly are you able to collect and receive data as it compare to your peers? And how much can you receive? Again, we're in a period of time now where it's easy to process data, it's cheap to store data. So go and fetch as much as you can, and even if you don't use it, you certainly can analyze some of it for use downstream.

Cleansing, and I talk about time reallocation here, if you can free up time for your data engineers to not be cleaning up and structuring your data, but really relocating what they do to more value-add processes. That can really set you apart. The data analysis piece, are you asking/answering the right questions? And with regard to AI, which is a very hot topic today, it's a lot easier than ever before to make use of some of these out of the box AI tools to allow you to analyze your data in the most efficient fashion.

You can now ask questions of your data loaded into some product or system and start feeding it questions, and it can give you back the insights. So again, you don't need to be that sophisticated data scientist to start asking and answering questions of your data. Collaboration, certainly you want to make sure that the right people are involved, because it's not just you and maybe the person that you worked with. It could be someone else on a different team, a client that might be providing you information. That collaborative approach is paramount to the success of this journey.

The stories and insights, again, I talked about data visualization, and I just want to make sure people are aware that these tools are out there that allow you to provide those real-time insights to people in the most efficient fashion. And then, finally, the data security. You just want to make sure that you're not going to lose out on something in the marketplace that could lose value for you, like a data breach, because that will set you back days or weeks in terms of getting your data back up. And I can't mention how important this piece is to the process.

I wanted to talk briefly, too, about the financial opportunities here. And I try to categorize it in one of two ways. Are you able to generate new revenues out of this? Or are you able to save money as a result of being literate in data and AI? I just wanted to give an example of each. I have some bullet points up on the screen which you can read through, but I'll give you an example of a client with regards to revenue generation.

I was working with a client that's in the events space and they started capturing more and more data about their wedding venues, who's attending, data points around food, food costs, how quickly they can deliver their services. And they just captured an abundance of data, which allowed them not only to create a better pricing strategy, but allow them to open up additional venues and know that they were doing it in the right way. So that was a way for an organization to become more data literate, start capturing more and more data, analyzing that data, and allowing them to enter into new markets.

And then finally on cost savings. That last bullet point I have there about fraud detection, I've worked with clients numbers of times where they are faced with fraud possibilities. And by going through this data literacy journey, we started capturing data, running it through a real-time dashboard to figure out, is there historical fraud? Or is there fraud happening in real time? And I worked with a college, for example, where we did just that, where we were able to go into the bursar's office, and start loading data into an infrastructure that allowed us to get the insights out that they previously weren't able to do.

So it's always important to start thinking about this and asking questions of where do you want to go? Do you want to start making more money? Do you want to save money? And these are things that you can certainly do as you go through this journey. Mechie, do you want to talk a little bit about competency levels? Oh, you're on mute again.

Mechie Nkengla:Absolutely. To sort of wrap up what Adam just explained about the opportunities here, let's think about this in context. The most successful, the most valuable organizations that we think of right now in the world, they're all data companies. And there's higher up on that competency level in terms of the maturity around data and AI literacy. So what are the competency levels really? And how is that something that we can track against?

The way I'd like to think about it is a collegiate level. So when you're in college, you have the 101 courses, the 201 courses, the 301 courses, and the 401 courses that are usually relegated to the seniors and postgraduate students. And those correspond to other titles, some call them, if you're 101, you are aware, you're data aware. You know That this is something that's of import, and you're inquisitive.

And if you're the 201, you're starting to learn. You're learning how you can talk with data, how you can integrate data, how you can ask the right questions with data. If you're 301, you're a practitioner, so maybe you're able to, if someone gives you some Excel sheets, you can pull up a graph, you can manipulate data in some fashion. If you're a 401, you're an expert at that level, you're probably a data scientist or engineer. You can write algorithms, you can write statistical models and whatnot.

So these are sort of ways that we tend to think about the competency levels. What is also important is, within an organization, everyone's going to be in different level. We're aware of that. There's expectation that not everyone is the same within an organization or needs to be the same, frankly. And so there are going to be a variety of competency levels, but in general, it pays to get stuck of where you are as an organization.

So to that end, let's move on to the next slide, which is another poll again. But it takes stock of where you think your organization is regarding these four competency levels on average, wrapped around. Again, you can click directly on the slide to select where your organization is.

Adam Karasick:And you don't lose points if you're a 101. I think it's important to have that recognition that maybe you're a little bit earlier on in that journey. So no shame in clicking 101 and no extra points for the overachievers either.

And I'll use this as another opportunity, if anyone has any questions, you can certainly push them through. We're keeping an eye on the Q&A.

Mechie Nkengla:  Let's give a couple more seconds and then we can move on.

All right. I think we're good. Adam, you want to show the results of that?

Adam Karasick:Sure. So it looks like more than 60% are at that 101 and 201 level. Everyone else is, potentially, above that. So a nice mix of attendees today.

Mechie Nkengla:And I want to call up that we do have some folks that are off the charts, so kudos. Wherever we fall, whether you're still yet to start on the 101 or you're way above that, we appreciate that. It gives you a sense of where you are and where you need to get going.

Adam Karasick:I'm going to walk you through now what we call the persona archetypes of this journey. I have up on the screen about five examples. There are certainly more that you can think about for your organization, but I wanted to show you the kind of process that we like to walk through with organizations when they go through this process of kind of aligning themselves to certain competencies.

So for example, we have a organizational stakeholder through to the data engineer on the top of the screen that you can read, but we have highlighted here on one slide, I have another on the next slide of what we might do to break down these individuals. So we have the business SME, the subject matter expert, up on the screen, and we have an example of four skills that this individual might have.

Now, the first one we have up on the list here is data skills. And I think that's the first thing everyone jumps to that they need to get all the way to the right, which would indicate that they have more skills. But the truth of the matter is I don't think it makes the most sense for every individual, every person in this process to have the most adept, most advanced skills in data. For a lot of people, this takes years to develop, and there are other attributes to the individuals here that are equally if not more important.

So the business SME is someone that we would say, "They really need that business acumen. They really need to understand what the organization's trying to do, what it's trying to accomplish." I know from my experience working with a lot of data analysts, data scientists, they may not have that level of business acumen. Maybe their data skills are a lot higher, but someone like the SME, we need them to really understand the business and communicate with them in the most effective way.

And then the last two we have up on the screen here, the soft skills, communication. These are often lacking as we see, but it's equally if not more important for you to be able to convey what it is you're trying to accomplish in the most effective way. Sometimes that advanced person needs to talk in a form or fashion that can allow them to easily communicate with all the other individuals in this process. So communication is always important.

And then I just jump to the next slide here, we look at the organizational stakeholder, which I think a lot of you attending today might consider yourself to be. And I have on the top slide here, the data skills, they're certainly not on empty on the left, but they're far enough over that you're at least aware of all the different facets of data and I. And that's important, because that will allow you to start asking those questions for all the other individuals involved, and really be able to immerse yourself in the data and AI space. Because without a little bit of knowledge there, you're not going to be able to start on this journey. And then maybe this individual, someone who's got very high business acumen and communication skills.

So I think the takeaway here on a slide like this is start thinking about who those individuals are that you currently have, who those individuals are that you would like to have. And then do you want to ask yourself, what are the skills that they currently have? And what are the skills that they should have? As you start crafting yourself through that process, you'll better position yourself to be able to go through this process on your data and AI journey, because it's not something you need to walk through on your own. This is a team-based approach and you want to align yourself with the right people. Mechie, do you want to talk about crafting a program?

Mechie Nkengla:Yes. Thank you. So we've talked about the different maturity of competency levels. We've talked about personas. Now let's talk about the heart of the matter, why a lot of you I imagine attended the webinar. How does this look like to craft a program for my organization? So when we talk about crafting the program organizations, there are a couple of things that we want to think about.

Data and AI literacy is not a one size fits all. Every organization has their own path and their own journey that depends on number of factors. The type of organization, so if I'm looking at a tech company, the AI and data literacy expectation in terms of competency will be different than a moms and pops fast food store, for example. And now those are two extreme examples, but nonetheless, you get the point.

And so it's always important to take stock of what is your vision for AI and data literacy for your organization? It always starts with that, right? We have this very nice quote here by Harry Kissinger about, "If you don't know where you're going, then every road would lead you nowhere." And so is this always important to craft out that vision, and use that vision as your North Star for whatever activities that you take. In the future, you have a guiding star in terms of where you want to go as that vision of what that looks like.

Now we begin with defining that vision. Once we have that vision, that becomes our why. Now, we need to do an assessment. We talked about the competency levels. It's important to measure where you are right now to define where you're going to go and by how much. We need to measure upticks, progress. So how do you measure your current competencies? There are things, there are assessments plenty around. We have one that we call the Data IQ, and it can be structured for a particular individual or for an organization at large. That gives you perspective of the progress that you're making.

Now you have the assessment for individuals and for the organization. Then the heart of the work starts, where we usually tend to work with the organization within the business, and we mark map out archetypes. So Adam walked through some examples of archetypes. It could be the C-suite, the leaders, the subject matter line of business experts. It could be more analytical. It could be administrative staff. It could be support staff, a variety of that.

And then the idea is for each of these archetypes that you've defined, so you're mapping job roles and titles to archetypes, once this is mapped, then the question becomes, what is the expectations of the data and AI literacy skills for each of these archetypes? Then there's work to sort of map how they expect someone in this archetype to be able to do such and such and such and such and such and such.

Once you have that all laid out, then we move on to the question of, how do I get from the baseline of where they are right now to my expectations of where they should be in terms of their competency? What program should I create that will help move them along that maturity curve? Those are all things that we think about. It's also important to think about this, as we're mapping out archetypes, one of the critical things to understand is there are different roles in the organizations and there are different for a particular purpose.

Everyone is specialized to do certain sort of tasks and do them well. We don't tend to speak the same language. I don't know how many of you have been in a meeting where you have some IT specialist and some business person and some legal person and some C-suite person, and they're having a conversation. And they could be using exactly the same word, but it means completely different things amongst everyone. It is critical that we understand AI and data because, frankly, it's embedded in everything that we do for more than one reason.

Yes, it's going to boost our career. It's going to make our lives easier, but frankly, it's going to help us communicate better. It's going to create that sort of common language platform that we can understand where each other's coming from. And when we communicate better, we can do projects better and make the organization a lot more productive.

So now we have our archetypes. Then I will develop this plan in terms of getting form baseline to the goals. Then what we do when we get there? Do we fold our hands and sit down? No. We think about AI and data literacy, it's not a destination, it's a journey. It's going to be continuous. So something where it is a race, so it's a marathon. You take a step and you reassess where you are and do the same thing and move on forward and forward. And the idea is progress, directional progress to where you want to go to map out in your vision, as opposed to we're going to do this course and then everyone's going to become data literate.

We all know that that's not never the case. It's not a one course or workshop that you give, and then everyone all of a sudden becomes more AI and data literate. So have a lot of patience in terms of how you map this out, because again, it is a journey and not a destination. And it's something that we're going to redo and redo over and over again, but we need to make that first step now. I see we don't have any questions. Oh, I think we have a few questions. Let's see here.

Adam Karasick: We do.

Mechie Nkengla:I see one question is, what a good example of AI in accounting professions, specifically the tax world would be? Adam, do you want to take that or shall I?

Adam Karasick:I have some thoughts there. And there's a couple other questions here which I'm sure we could spend a lot of time talking about. I'll give you my quick thoughts around AI and tax. I know with Intuit, with TurboTax, and I know H&R Block, they sit on a mountain of tax information about filers, individuals, corporations. What AI could allow you to do with a trove of information like that is figure out, with a new tax filing, what may or may not be missed? What filers of that type might be doing or leveraging? And how that individual or how that corporation could benefit?

So in much the way AI loads in a lot of information and scans it and looks for patterns, they could use data AI component of a tax filer could use that amount of information to assess how taxes are being filled out, how they're being captured. I think there's a lot that is certainly possible. I think it's important to figure out if you want to go on that journey, what is it you might be trying to accomplish? Whether it be reducing the risk or finding additional deductions, for example. But if you have a lot of other similar data, that's where an AI tool can start to analyze all that information to provide patterns. I mean, that's really what AI is trying to accomplish. Mechie, I don't know if you have any additional thoughts there.

Mechie Nkengla:No, I think you captured it right, and there are lots of use cases we can go through and we can share that information with you later. Feel free to ping any of us, and we can show you a host of examples of how AI and machine learning are used in that accounting and tax specifically.

I want to get to the next question, which says, with a recent rise in ChatGPT popularity, can you talk about the benefits of good prompting and how a user can be more successful than another simply based on their ability to ask the AI good questions? I love this question. Thank you to the audience member that posted this. Well, our conversation today is really not about ChatGPT, but there are a couple of points I want to bring up here.

The first is when we think about ChatGPT, we should take a step back. It's not just an AI, it's a specific type of AI. It's one that's based off a model called large learning models. And the idea is it basically learns humongous massive amount of information, and it behaves based on its learning from that massive amount of information. So it's really bounded by what it learns from that. So if you're going to ask a better question, my response would be more specific. The more specific you are in your prompting, the better the response.

However, it's worth noting that, again, there is a limitation, it's based on learning. So if you're going to have this solution give you a response and answer based on what it's learned, it can only respond to you from that context. I don't want to spend too much time here, because this is a sort of whole topic that we can do a whole other webinar on, but I hope I answered your question.

All right, let's move on to the next slide. We have another poll. We're sort of wrapping this up and I want to talk about some key takeaways here. But what's your biggest takeaway thus far in terms of what we've shared and conversations we've had here today?

So please again, feel free to click on any one of the points directly.

Give another second. Adam, while people are voting, let me ask you one question that came up, and this was-

Adam Karasick:Sure.

Mechie Nkengla: ... a great one. It goes, please describe the components of cleansing. I think it's referring to cleansing as part of the work that needs to be done to prepare the data.

Adam Karasick:So there's a couple different facets of data cleansing. Again, with some of the other topics that have come up. I could probably talk for an hour about data cleansing, but I think there are aspects to cleaning up new data as it's captured and entered into some sort of system or software. There's a component of this where you're cleaning up data that's already in your system.

Perhaps you've got some field that's being captured where it's just freeform text and your goal is to take that information and structure it. You can think of descriptions, user feedback, where you want to turn that into some sort of value. So that's a data and engineering component of this to clean up that data. Then there's a lot of one-off ad hoc projects where an analyst might be asked to take data from different systems, bring it all together so that you can analyze it in one spot or in one way. The sophistication to be able to merge and marry all that data together is an area of data cleansing that is often overlooked and underappreciated, but that is an area of data cleansing that is extremely valuable for organizations.

So it really comes down to a question of what are your issues? What are you trying to accomplish? And what tools and types of people are out there that can help you with that process?

Mechie Nkengla:Thank you. And as we can see here, we have about... I think the last thing I said, journey, not a destination. Yes, that's quite important. "Yeah, we need to get on that bandwagon right now and it's critical to my career growth." And we have a few, we're awesome. So thank you for that. Adam, you want to take on the next slide, please?

Adam Karasick:Sure. So we want to talk a little bit about some takeaways from this presentation. Again, I mean, we mentioned it at the beginning, which is that there's a lot that you might encounter as you start taking these steps forward. But if you keep your scope small and narrow, if you start going after that low hanging fruit, that will allow you to create the momentum that you need to really move forward.

So that's an important thing, which moves into this slide here, which is that there is no one size fits all approach for this. Different organizations have different goals, you have different resources available to you. Some don't have the money, some don't have the employees or the staff to make it happen. I know when I started looking into AI several years ago, I was overwhelmed by what I see on TV, in the news, in movies. I think the AI applications that I see are often those very large, massive solutions that require time investment, a lot of people.

There are AI opportunities out there for you now that are much smaller scale that can provide you some immediate value in a matter of weeks or months, not years, not millions of dollars. So don't be overwhelmed by the solutions that are available to you. And I think it's extremely important, too, that you really assess your current state and figure out what's achievable in your short term and long term, and then start asking the questions amongst all of your stakeholders and individuals who are going to be the decision makers about, "Where do you want to go next?"

So it's extremely important that you realize that this is not about something that's so hard to achieve, so hard to really get into a successful state. For you, it might be different from an organization that's similar to you. And I just want to move into the next slide here, which talks about starting your journey and the importance of it. So this is a slide that I pulled from Our World in Data, it's a website that has a lot of data that's great for analyses, and it talks about the rate of adaption or adoption rather of different technologies.

So there were more that are available, you can go play with it if you'd like, but I have up on the screen five. And you can see as time has progressed, as we've gotten closer to the current day, the rates of adoption are exponentially faster. So the first one I have up on the screen are the toilet, which back in the late 18 hundreds was slow to adopt. And that might be because of the lack of economies of scale, maybe people in certain countries didn't even know that this technology existed.

And then as we move forward to the car, the color TV, it still took a number of years for us to reach 50, 60 plus percentage of people who had that technology. But then you see desktop computers, the cell phone, those rates are almost vertical lines. And then we talk, I know it's been a hot item in the news recently with ChatGPT, a lot of the AI and data technologies, the rate of adoption is extremely fast.

So I think it's extremely important for all of you to be thinking about these things, because if you are not, it's likely the competition is. And I think it's important that you start this journey, as we keep saying, becoming literate so that you know what questions to ask, you know who to call, you know what you want to do. Mechie, do you want to talk about communication?

Mechie Nkengla:Let me see. Thank you. Yes. So we started out by defining both AI and data literacy, a key part of it is the ability to communicate and talk about them, whether it's conceptually or within context. One of the things that I wish I had learned and not to age myself, when I started my career many, many decades ago, was to understand how critical communication is. So oftentimes we talk about how it's important to know your work and be able to do it, and yes it is. But beyond all of that, it is critical to communicate.

Communicate what you know to influence your peers and drive solutions and make impact and tell what's going on. So here's a great slide that gives you an example of the impact of communication here. We have data and we see how when it's sorted, when it's arranged, and then when it's explained with the story. Like Brené Brown says, "Storytelling is about putting soul into data." And there are different ways that we can go about communicating with data.

This itself is a whole other webinar and old course around visualization and communicating with data. And this is a skill set that's important to anyone. Whether you're an analyst or a leader, communication's part of what you do. You're communicating your results, you're communicating your strategies. And when you're communicating with data, it makes them much more foundational, much more impactful. And if you weave a story around them, then people relate to that. And we can probably schedule another webinar on some of the key things.

One example of a good way to communicate with data is give results in three. And I'm sure there are lots of consultants here that would agree with me. The human brain connects with a number of three rather than give two or four, give three examples, typically, usually connects with human.

Another key example is working in a pyramid sort of manner, where you're starting with your key point, what's the conclusion of what you're trying to make? State that, and then work downwards to sort of explain why you write a conclusion, and then back it up with facts and figures that speak to that point. That's another great way that sort of pushes through in terms of communicating.

And I just told you the story around the villager's well. I bet you guys are going to remember data colonization because of that story much more effectively than any other things I probably said here today, because I used the story to illustrate that example. So again, communication is key, and it's one of the single most critical skill sets that defines the success, your success. Next slide.

Adam Karasick:I think the other takeaway I want everyone to have here is that AI and data literacy can really be a superpower for you both individually and within your organization and amongst your competition really. I know when I started my career, I was often that go-to Excel person. And for me that was kind of my kind of superpower, I would call it, as I was traversing through my professional journey.

But I think if you make your organization more than just the service you provide, more than just the product that you might sell, if you can show customers, if you can show your employees that you're an organization that really leverages AI and data analytics, that it's a part of the value that you're providing for end users, it's a value that you're providing so that your colleagues can really feel like they're empowered to make the best decisions on a day-to-day basis, that's going to enable you to really set yourself apart. And it's never too late to start this journey. It's never too late to become adept at a lot of these core competencies.

So again, I know for me, I first felt overwhelmed by the whole universe of AI and data analytics, but I think you have to start and start taking those baby steps in order to figure out where are you going to separate yourself from your peers? And if you think of it as a superpower, I think it will certainly enable you to want to learn and do a whole lot more. So we're going to take some more questions before we wrap up. Mechie, there's a question here about AI and machine learning. Do you think you want to take that?

Mechie Nkengla:Yes. Let me read the question out loud. So you've talked in general about AI, please compare and contrast AI and ML and what I need to know about each. So let me start with the generalization. So an intelligent device, or let's say system, uses AI to think like a human and perform task on its own. And this could be perception tasks, this could be thinking tasks, this could be communication tasks with the natural language processing, all of the above. And that is what an AI system is.

Now, machine learning is how the computer system develops that intelligence. So while these two concepts are very related and connect to each other, they're different. And speaking of data literacy, one of the things that we find, and you've been in meetings, people use those term interchangeably and they're not the same thing. And so think of machine learning as how that system develops intelligence, and AI is really the idea of that system behaving like a human, behaving with human intelligence.

And I can get a bit more technical, but I think maybe we leave it a little bit like that and we please feel free to ask me a question. I can go into more deeper technical explanations later. But those are two ways I would like for you to think about it. And another example, so a good example in terms of story, if you remember when you clock in, so you get into your building and you have a system that scans your face and says, "Oh, Mary has arrived at work." And so it recognizes that you've arrived.

Now where is the AI and where is ML in that case? Think of machine learning as that algorithm, that heuristics that has learned to capture you in different phases, whether you have your hair up, you had a short hair, long hair, you have different clothes, you have your winter coats or not, and still be able to recognize that this is Mary. Now that is how it develops intelligence, and continues to learn that you're the same person and track into the system as someone that is approved to be in there.

The AI aspect will be where, now, understanding that this is you, what actions should it take? It basically sends in another process that says, "Mary is in," or maybe allows you access to certain floors and create things that are specific to you. That's the AI part of it. I hope that helps, explains a bit better.

Adam Karasick:You also have a question about starting the educational part of this self-studying learning. Mechie, do you have any thoughts about where people can go to take some courses or maybe some methods around education on their own?

Mechie Nkengla:Yeah, so one of the things that's going to be provided is as part of this platform, there's going to be.a list of resources that we can get to go to get a lot more information in general, and you can contact Adam or I for more information.

But in terms of getting started as an individual, again, the same concept roughly applies. What is your why? What's your vision? How data literate or AI literate do you want to become? Do you want to become someone that runs models and builds algorithms? Do you want to be a business person that can effectively understand what AI and data is and communicate with it and make business decisions from that?

Those are, again, different sort of competency levels that you're striving at. And then from that, create a blueprint. What does that mean if I can communicate with data? And again, you start by building out this basis blueprint of what it means for you. And you start by taking the most basic education classes. We're going to provide some resources for you that you can go through and play around in terms of how to become more AI and data literate yourself.

Adam Karasick:And one thing I want to add, too, I often get this question and people think that there's some sort of certificate program or all day training that they can go attend, they'll go and then they're going to walk out of there and be totally proficient at this thing. But the truth of the matter is that when I went through this journey, what really was helpful for me was making it a culture for myself of just self-study and education over time, sprinkling in little trainings, little videos, even podcasts, and making it a part of my self-development.

So if there's a certain area or competency that you want to develop for yourself or for others, I strongly recommend that it becomes something that you sprinkle in and make it a part of your day or your week over time. And eventually it really will develop in a much different way than say, walking into a four or eight hour training and trying to absorb as much as you possibly can in one sitting. So that can certainly make a sizeable difference, in my opinion.

All right. Well, I think we're up at two o'clock, so if there's any other questions, you can certainly send them in. But I think we're going to wrap up here.

Mechie Nkengla:Yes, I want to thank you guys again for coming. It's been a pleasure to be here. And I hope Adam and I can help you guys now and in the future in terms of your AI and data literacy journey for your organization.

Transcribed by

What's on Your Mind?

a man in a suit

Adam Karasick

Adam Karasick is a Senior Manager within EisnerAmper’s Digital Solutions practice, serving as a team leader for the data analytics and business intelligence unit.

Start a conversation with Adam

Receive the latest business insights, analysis, and perspectives from EisnerAmper professionals.