Everyday AI Podcast – An AI and ChatGPT Podcast - EP 337: What Happens When AI Works? Tackling Responsible AI.

Episode Date: August 15, 2024

Win a free year of ChatGPT or other prizes! Find out how.No one's talking about this when it comes to AI. What happens when it works? ↳ When we save all that time. ↳ When productivity goes through the roof↳ When jobs maybe get.....easier. Diya Wynn, Responsible AI Lead at AWS, joins us to discussNewsletter: Sign up for our free daily newsletterMore on this Episode: Episode PageJoin the discussion: Ask Jordan and Diya questions on AIRelated Episodes: Ep 323: How AI Is Changing Workplace ProductivityEp 302: 5 Laws for Success in the AI EraUpcoming Episodes: Check out the upcoming Everyday AI Livestream lineupWebsite: YourEverydayAI.comEmail The Show: [e-mail] with Jordan on LinkedInTopics Covered in This Episode:1. Allocation of resources and the role of AI2. Responsible and intentional AI practices3. Complexities of upskilling and reskilling4. Implications of AI on a grand scaleTimestamps:01:50 Generative AI impact on high school students.05:40 About Diya and her role at AWS10:28 Inclusive AI promoting fairness and accessibility.13:47 Augmenting human capability with technology for efficiency.15:31 Comparing AI to automotive revolution, impacting jobs.19:50 Resistance, failure, and education for responsible AI.23:27 Skills need constant updating for technological advancements.26:08 Responsible AI as an organizational culture structure.32:21 Resource allocation may impact access to healthcare.35:42 AI levels playing field for diverse learners.38:59 AI as bridge, not barrier, empowering all.Keywords:Diya Wynn, Jordan Wilson, Everyday AI Show, AWS, Medicare, Medicaid, AI implementation, disparities in access, Artificial Intelligence, organizational structure, AI practices, education, banking services, hiring processes, video production, systems testing, upskilling, reskilling, responsible AI policy, technology advancements, job displacement, job creation, biases, fairness, workforce changes, World Economic Forum predictions, AWS responsible AI strategy, heart disease prediction, industry events, community engagement.Send Everyday AI and Jordan a text message. (We can't reply back unless you leave contact info) Start Here ▶️Not sure where to start when it comes to AI? Start with our Start Here Series. You can listen to the first drop -- Episode 691 -- or get free access to our Inner Cricle community and all episodes: StartHereSeries.com Also, here's a link to the entire series on a Spotify playlist. 

This is the Everyday AI Show, the everyday podcast where we simplify AI and bring its power to your fingertips. Listen daily for practical advice to boost your career, business, and everyday life. Meet Firefly AI Assistant, now live in Adobe Firefly, the all-in-one creative AI studio. Just describe what you want to create and the assistant handles the rest, orchestrating multi-step workflows across Photoshop, Premiere Express, and more in one conversational interface. You direct the outcome. The assistant accelerates execution. What happens when AI works, right?

We spend so much time and thought and energy into implementing generative AI and large language models into our companies, chasing these golden claims of being more productive, 30%, 50%, 70%. But then what happens when we see those productivity gains or those efficiency gains? What happens with the humans? Do we sit around and wait for a new project? Do we start working in other departments? I don't know those things, but it's questions that I think about all the time. Luckily, on today's episode of Everyday AI, we have an expert joining us from AWS, who that's her role. And I think she is one of the smartest people out there when it comes to responsible AI.

So we're going to be tackling that together today on Everyday AI. What's going on, y'all? My name is Jordan Wilson and I'm the host and Everyday AI. It's for you. It's a daily live stream podcast and free daily newsletter, helping us all learn and leverage generative AI to grow our companies and to grow our careers. So if that's you, maybe you're listening for the first time. Thank you for joining us.

If you're on the podcast, make sure to check out your show notes for a link to our website, a link to the live stream where you can come in and ask questions after the fact. So make sure to go to your EverydayaI.com. Sign up for that free daily newsletter. We will be recapping today's interview and a lot more in our newsletter. that goes out, as well as make sure to check out that thanks a million giveaway campaign. We have going on celebrating a million downloads here at Everyday AI. All right, before we get into it, let's quickly tackle what's going on in the world of AI news.

So a federal judge has allowed a key copyright claim against AI image developers to proceed. So a federal judge has advanced the copyright infringement case against AI image developers like stability AI, mid-jurney, and the online art community deviant art, marking a significant step in a high-profile legal battle. So U.S. District Judge William Orick has allowed the copyright and trademark infringement claims to move forward while dismissing claims under the Digital Millennium Copyright Act, DMCA, and unjust enrichment. So the lawsuit was filed by artist Sarah Anderson, Kelly McCurnan, and Carla Ortiz,

who alleged their work was used without permission to train AI models. Stability AI, Deviant Art, runway, mid-journeys, and others have yet to comments on the ongoing case. So pretty, pretty big deal there that this, a judge, a federal judge, is allowing this case to move forward. All right. Next, a generative AI usage in math prep study is showing that using AI, generative AI in math prep is actually leading to lower exam scores. So a new study from the Wharton School reveals that high school students using generative AI for math exam preparation perform actually worse on actual tests compared to those who didn't use the tools while prepping. So here's the significance.

You know, AI optimists, you know, are obviously envisioning a personal tutor for every student, but the study highlights that that might not be the case as there's challenges and potential drawbacks for AI-driven learning. So some schools have banned generative AI tools in large language models, while others are really pushing them and permitting their use with disclosure. So as an example, Khan Academy's Saul Khan piloted a generative AI tutor last year called Khan Migo, super popular in the educational system, aiming to help students solve problems rather than just providing answers. So yeah, that one's an interesting one and we'll be keeping an eye on that. Last but not least, World Labs has reached unicorn status in just four months. So World Labs, a startup founded by Stanford AI professor, Fifi Lee, has a chance.

achieved a valuation of over $1 billion within four months of its founding. That's wild, y'all. So Fifi Lee has been given the name the godmother of AI. So according to a report from TechCrunch, the latest financing route led by N.EA raised $100 million and significantly increased the company's valuation from $200 million in April to now a billion dollars. So the startup aims to develop AI models capable of accurately estimated. estimating the three-dimensional physicality of real-world objects and environments,

potentially revolutionizing industries like gaming and robotics. Fifi Lee is often referred to as the godmother of AI. She highlighted the importance of developing machines with human-like spatial intelligence in a TED talk earlier this year. According to an investor familiar with World Labs, the company's approach could reduce the need for extensive and expensive data collection, which is currently a significant hurdle for many AI applications. All right, we're going to be having a lot more on those stories inside of our newsletters.

So make sure you go check this out. But let's talk now about the topic for today, which is one I'm constantly thinking about because if I'm telling you the truth, I'm very lucky. I get to talk to the smartest people in the world. And this is one of those issues that I think is hard to tackle, right? What happens when AI works? How can you responsibly, you know, not just roll it out in your organization, but responsibly handle everything that comes with it.

So I'm not by myself today. Luckily, I have a guest. So I'm excited for this one. Please help me welcome to the show. There we go. We have her, Dia Win, who is the responsible lead at AWS Amazon Web Services. Dia, thank you so much for joining the Everyday AI show. Thank you for having me, Jordan.

It's good to see you. Oh, I'm excited for this one, y'all. I've been wanting to get Dia on the show for a hot minute or a hot couple of months. Dia, can you just tell everyone to live? little bit about what your role entails at AWS as the responsible AI lead. Sure. Well, I'll tell you all something interesting. Actually, it's just a couple of weeks ago. I am in a new role. So just move over into responsible AI policy, but which will look a little different in that, you know, I get to work with folks that are in legislation or working on legislation around responsible AI. So it's

an expansion of the work that I was doing previously. But, you know, I had the opportunity to start our customer-facing work on responsible AI. So essentially, one of our first internal practices solely focused on thinking about the risks and applications of AI and doing what we do well at AWS in terms of helping guide our customers and support them, partner with them as they develop and build on top of our services. And we use the very similar model to what we do, with like the cloud in that we provide a structure to be able to identify best practices, support them in terms of understanding

and growing in their ability to be able to leverage the cloud well and maximize their benefit. While we use that same sort of model in terms of thinking about responsible AI, how do we support them given the hundreds or maybe even thousands of a thousand data scientists that we have internally throughout the business, working in various areas,

the folks that we have in policies, the work that we're doing in standards, you know, the experience that we have working with our customers who are building, as well as building for, you know, our platform or our company, Amazon, bringing all that to bear, as well as the ongoing research that we've been doing to be able to help them think through the areas of risk and put in place kind of best practices that would help them address those risks, minimizing the potential for impact or negative impact, and maximize the benefit. and hopefully the benefits, not just their bottom line, but to all.

And, and, and, I really do want to get back to what your role looks like a little bit more at AWS, because I'm curious. And I think that, you know, our listeners can probably gain a lot from, from understanding how a huge company like AWS handles responsible AI. But I actually just want to take a second and fast forward to the end and ask the big question. So, you know, what happens when AI works? What happens when, you know, companies are maybe finally realizing those 20%, 30%, 50%, you know, increases in productivity and efficiency?

What happens then? What happens with, you know, employees that maybe have way more on their plate than they, or sorry, way less on their plate than they did before? Yeah, well, no. I actually, and I said this to you, I love the topic because, you know, oftentimes in my work, I'm focused on like the risk and helping people understand their risk. and understand the unintended impact, ultimately, in order to see the good when it works. But I have the opportunity to look at this much more critically from an area of risk and elevating those risks as well. So when it works, I think it's ultimately a bridge, not a barrier, for people to be able to reach their full potential and for us to drive towards more equitable

outcomes with the technology and build a more equitable world, right? That's sort of my envisioning of what this looks like, you know, when AI works. And that means that we have, you know, systems that are not necessarily replacing people, but are working and augmenting human capabilities so that we can be our best versions of ourselves. As you mentioned, when we are, you know, I just understand and know that there are many of us

that are overworked and but have systems to benefit, you know, and to leverage in terms terms of, you know, being able to do everyday sort of tasks, but also allow us the space to be able to, you remember that. There was a book that was like the, what was it? I think the, what was it, it wasn't the four-hour work day or so, the four-week, four-day work week or something like that. I forget the book. Someone bought it from me, and clearly by me forgetting the title, you know, that I haven't mastered that. But the idea was that we could do. Four-hour work week. Four-hour work week. Okay. So that's what it was.

Right. But the idea, like, imagine actually being able more of us to being able to live into that and be less stressed and be able to maximize our creativity and capability. Imagine a world where we have, you know, folks that, you know, typically would be underserved and marginalized and underrepresented, actually lowering various entries so that they can, you know, be part of and included in systems. Right. I think, you know, systems and technology that address some of the biases and promote fairness and decision-making processes, whether that's in hiring or criminal justice or where resources are allocated. When it works, it's, you know, we have the right kind of inclusion in place so that all are being considered kind of the same ways that we thought about in earlier days of thinking about disability and disability. ensuring that we're building, you know, for accessibility that benefits us all, right? All of that, I think, is when AI works and when we're doing the intentional things to minimize the areas of risk and impact or the potential impacts that the systems can have and that we're all very familiar about and concerned about in terms of, you know, its development and advancement.

Yeah. And, and, Dio, I'm also curious, what about on the flip side, right? you just painted out the utopian side of the coin, right? When we have the more four-hour work week and we have the ability to work on, you know, in driving equitable outcomes. And yeah, that's 100% possible in a future with generative AI. But what about the more dystopian side of that coin when we say what happens when AI works? Because, you know, a lot of companies, you know, what they're doing is, you know, we saw it from, you know, I think IBM, Intel, you know, laying off a lot of employees and saying, hey, we're going to

focus on AI, it's going to drive efficiency. So what about the more dystopian side of that coin of what happens when AI works? Because a lot of people are scared of that when AI works. Right, right. So it's interesting, right? Because I've looked at this, what happens when it works when we're doing the things, right, to build into and when responsible AI, the area focus that I have is part of the way that we work. Because part of my envision and those that are in spaces like mine, or roles like mine, think about responsible AI being the way in which we build, that we ultimately want to move to a place where we think about, like, and are intentional about who we include and whether there's value alignment and what we're doing from

transparency and fairness. That is the world in my mind where AI works, where we're building with that kind of intentionality, you know, across the life cycle. So what you're talking about the reverse of this, when it works, right? and the ways that we're envisioning being able to do new things, have greater efficiencies, and businesses are looking at, well, if I have greater efficiencies, then I can, I need less people. Or I need, or we spend less in certain areas. That means that we have the opportunity to be able to be optimized and then can invest in other things.

I think that what we talk about when we work with companies, and even internally is, one, you know, the focus on human beings, right? And not necessarily looking to replace human beings, but that we're augmenting capability, that we look for opportunity for humans to work with, systems and technology to be more efficient, to remove the undifferentiated heavy lifting, kind of the way we talk about the cloud as well. And then we get to sort of optimize or be focused in the areas that really bring out our unique value. But that, again, doesn't happen without. intentionality. So that means that we have to be thinking about the ways in which if we're replacing

work or that people are now working with systems, that we actually have to focus on the kind of education of skilling and rescilling that is necessary to be able to one help people understand how their work shifts as a result of that and then prepare people for pathways to be able to move into new opportunities or new areas of work or new roles that are being created that weren't previously existent. All of that has to be part of the intentional focus that companies, you know, embrace the technology with this sort of mindset, this organizational change, this structure in place to be considering that. And I think it's not just a responsibility of, you know, technology companies or those that are employing AI, but that's also, you know,

part of what we expect in government as we look at and NGOs and others that are looking at workforce programs that we're thinking about that as well. Because with any technological advancement, any major error, we've seen e-shift from the kind of work that we had, right, to the work that we're now doing today. I heard someone mention this the other day, and they likened this move with AI and generative AI being similar to like the way that automobiles changed the face of everything, going from horse and buggy to having like automobiles that changed our supply change that changed how we got mail that changed the work that we do that change industry right it wasn't just that one area it trains how we gather with individuals right like all of this was a massive adjustment

and it didn't mean that we didn't have people that were driving horse and buggy or or like the elevators of all that we don't have people that are helping us, you know, that we're elevated operators. They're now different responsibilities that we have. And we have to understand that with all of that technological change, there will be differences in terms of the work. But then we have to prepare people for the work. The World's Economic Forum said that, I think it was what, 70 million jobs would be displaced, but 80 million would be created. And it's not necessarily to say that the people that were in the 70 automatically moved to the 80, which is why that intentionality around re-skilling and upskilling is essential, right,

to be able to hopefully create pathways for people to be able to move into the work of the new and to the new opportunities that we're seeing in the future. Yeah, and I think that's what I think when we talk, or at least when I think about responsible AI in the future, that's one of the places my mind ultimately goes. And hey, as a reminder for our live stream audience, if you have a question for DIA, you know,

it's not every day you can ask a question for the, the responsible AI lead at Amazon Web Services, but get your questions in now. But let's focus on that a little bit here, DIA, in this concept of, yes, there's going to be, I think, massive job loss, but there's also going to be massive job creation, right? You just mentioned the study there from the World Economic Forum, that 70 million jobs would be impacted,

but potentially 80 million jobs created. So, you know, how are you even approaching this at AWS? And how should companies be looking at this process of upskilling and reskilling, because from my perspective, I think it's personally hard, because in some of these instances, it might not be a one degree or a five degree shift in terms of where you might be upskilling and reskilling someone to, but it could be a 90 degree or 180 degree change. How can companies be doing this responsibly? Adobe just introduced an entirely new way to create, bringing the power and precision of its

creative suite into one conversational experience. Meet Firefly AI. Assistant, now live in the Adobe Firefly app, the all-in-one creative AI studio. Powered by Adobe's creative agent, Firefly AI Assistant lets you start with your vision, just describe what you want, and shape the outcome as it takes form with the assistant. The assistant orchestrates multi-step workflows, drawing on 60 plus pro-grade tools across Adobe Creative Cloud apps, including Photoshop, Illustrator Premier, Lightroom Express, and more to help bring your ideas to life. You can also get started with creative skills, a growing library of pre-built workflows for common creative tasks, like batch editing photos, creating mood boards, portrait retouching, and creating social variations.

Every step the assistant takes is visible so you can refine, redirect, or take over at any time. You stay in the driver's seat as the creative director. Adobe Firefly AI assistant now in public beta. See it today at firefly.adopi.com. Yeah, I think I think that that is a great point. And, you know, I say this, like upskilling, resale, feeling like it's easy and it's not. And I even remember when we were having conversations, again, I liken this to the cloud, right? We talked about, like, the cloud shifting the work. We have people in data centers that were focused on managing infrastructure, right,

that now perhaps we're going to have a different responsibility and wasn't going to be the same when we have systems in the cloud and not necessarily in the data center that they were physically responsible for. And that was a shift of work. And we had folks that were willing to make that shift and others that were not, right? And so that's part of some of the complication with this as well. There's going to be a degree of resistance. There will be some degree of failure, all of what's to have to be work through in order for us to shape and reshape like the opportunity for people in the future. But I think that's part of the work. And again, not just for those that are intact or those that are, you know, building products, but also for sort of community organizations, other institutions, as well as

government to have a role in the place at play in. But like what are things that we do? I think one of the things is one educating people on, you know, the value that the organization is placing in AI and what their commitment is in terms of using AI and being transparent about their policy as far as responsible development. I think that's one thing that's critical. And we were talking about this in the way that we think about some of the most organizations have compliance training that happens every year, right? Like you have to understand security and then we need to understand like data classification in the way that data is used and we need to other, we'll also do some other sort of test like a requirement required test to understand like what is sexual harassment,

etc. Like those kinds of annual compliance required training could be an opportunity and we're certainly some that are using that model to roll out, like, awareness and commitment to an understanding of our use of AI, where it gets applied, what our commitments to the development of that is. We started last year through Amazon-wide initiative under our inclusive tech team, like actually training sort of broadly on AI and responsible AI. And then making sure that we also have like role-based or AI education and training that is aligned to people at different stages of their life cycle. So what is necessary and essential for someone to understand if they're building and responsible, like from a data perspective, their data scientists or an engineer is different than what we might want, you know, someone to consider as a product manager or in IT. And so having role-based or that kind of training that's aligned.

across the board. And then we've done things to make AI education sort of broadly available to our employees as well externally. So our skill builder platform is one of the learning and development platforms that is also open to the public. And on that platform, we have a lot of low cost as well as free training and education that will provide AI education to folks like thinking about it and strategy and business as well as, you know, specific skills in terms of those that are looking at Python development or, you know, using one of our services. And so that's made available to our resources internally as well. And so I think that we're thinking about, you know, critically, thinking about what does that mean as we look at teams. And that has to be ongoing or look at

leveraging and adopting the technology throughout our organization. But that also has to be, has to be an ongoing work, right? Because what we're looking at today in terms of roles and where AI is being used will very likely be a little different than tomorrow and having that sort of constant iteration and reviews that we can make the adjustments necessary, pivot and meet the demand is necessary. And one other thing I'll talk about this in terms of skill

and re-skilling. I think it's important as well in education context, right, that we are making the shift, right? There was a study, I believe IBM was one of the recent ones that talked about the half-life of skills. And in the last sort of evaluation, there was this notion that the half-life of skills was somewhere around two years.

I would venture to say that that is decreasing as well with the rapid advancement of technology, which means that we have to be ongoing and constant learners. And then that requires as well the way that we look at skills and learning to be different, both in sort of broader education contexts like traditional education as well as other programming, because we need to be able to meet the demand as we are shifting with technology. Now new skills are required. And the way that we look at education has to shift to be able to more of a skills-based sort of understanding. And we've

been talking about that for a while, but I think now what's happening with AI and the shift and roles and the shift in work, right? This is elevating that need and demand, what also means that we have to shift the way that we look for people, skills, and resources as well in our companies, right? Not just, do you have a four-year degree? And while there's necessity for specialization, right, because AI can do some of the lower level things and now we need greater specialization, I think there is still also a need to look at this skills shift that's required and looking at transferable skills and where we can get skills attainment and other opportunities outside of traditional school and elevating that as a means for bringing in resources

that are necessary, you know, throughout the organization. All of that full ecosystem shift I think is necessary. And all of it hasn't progressed as quickly as the technology is. So in some ways we're playing catch-out to shift our paradigm and I think near around us. Yeah, I think that's a great point. the concept of the half-life of skills or when skills start to use their value because, yeah, historically, it's been going down, right? And it's at the lowest point ever, which I think is why, you know, some of those points that you just made about, you know, companies providing education, ongoing training, et cetera, is extremely important. This question

from Cecilia here, I think is great. And I'm going to add on to it as well. So she's asking, how are you communicating the concept of responsible AI within AWS? But then I'll also, add on top of that, Dia, how are you not just communicating responsible AI, but also how are you communicating upskilling and reskilling? Because I think how companies talk about this, right, especially when they're implementing AI for the first time on a large scale is extremely important. So how are you doing those things? Right. So I think, well, first of all, let me just say, it's not just me. So there are a ton of people, fortunately, that have responsibility, you know, for like a responsible AI, how we're building and developing our services. One of the things I think that

primary is one thinking about this as like an organizational structure. Responsible AI isn't just principles or or tenants that are being followed, but it is, you know, intifle into the way in which we develop, build, and we think about AI. So I think that's one of the key things that this is an organizational structure. When I talk about responsible AI, it's a it's an organizational structure that establishes a culture of responsibility that helps us to incorporate people, process, and technology to be able to reduce risk, unintended impact, and marriage. maximize benefit. And so when we think of it that way, then there are our, you know, organizational change mechanisms that we put in place in order to help, you know,

increase understanding and awareness, but also bring people along that journey. Then we also have to have strategic alignment in terms of our leadership, right? And our leadership is a huge part in terms of driving the commitment to responsible AI, not just in Word, but also in action throughout the organization. So I think that's part of it. We have a responsible AI strategy, right? That informs and drives of what we do in terms of our commitment to our customers, as well as in development of our services, and the commitment long-term to how we are investing in the next generation of technologists, of engineers that are going to diverse to help shape the technology,

as well as the ongoing research in terms of responsible and ethical development of AI. All of that strategy underpins how we are, one, communicating about that internally. And then we have things like we, because it's a huge organization, we have to have like internal road shows. We have, you know, talks that will discuss that. We have training, you know, that incorporates it. We are talking about that in our organization and team meetings and on, you know, main stages and our off-sites as well as in, you know, broader venues. And you all are starting to see that as well. And, you know, in some of our, you know, major events like reinforce and reinvent and reinvent our.

reinforces our security event, reinvent our annual customer conference. Those are things that are on the main stage reflecting our ongoing commitment to responsible AI, talking about it so that others are awareness. And then, I mean, fortunately, I and a number of others get to do public things where we engage with the community, public, you know, industry events talking about responsible AI, right? That also helps increase awareness not only internally but externally as well. You know, one thing, you know, on this topic of what happens when AI works and looking at responsible AI, because I'm sure it's not an easy role for you and your colleagues at AWS DIA, right? Because this is huge, right?

This is, this is, it's topical, it's philosophical, it's, it's moral ethics, right? Having to juggle all these things. What's the one concept of, you know, what happens when this works? What's the one thing that maybe still not necessarily keeps you up at night? But what's maybe the one key or one, you know, specific aspect of responsible AI that still has you focusing and in scratching your head trying to figure this out? All right. So I would say, like the first thing is, and I said that there are, you know, teams of folks. I'm not, I'm not an only, the only individual that is to do this.

So the one thing that I would mention is that responsible AI is everyone's responsibility, right? It's not just reserved to the technical folks or the person that has a title with like the responsible AI lead. But ultimately, we all have to buy into this notion of considering the things that are important that we introduce into the process, transparency, fairness, security, safety, and our development, implementation, and even through a long time like deployment, what happens in the world in terms of the technology being developed. And that's important because we can do all the things in design and implementation. But if we don't think about like how that gets lived out and delivered in practice, then we still will miss the mark. And this also connects to why the thing that I think about. So I'll use this in the context of an example. We have a customer that is able to now leveraging AI 15 months prior to clinical manifestation,

be able to determine the likelihood of this individual getting heart disease. Major, major sort of move in advancement in technology. That's important because heart disease is the number one killer of all demographic, people of all demographics in the United States, right? So imagine the idea that we could actually save people's lives, provide preventative care to keep people from dying every day, every hour from heart disease. Major, major sort of advancement. But you do all the things to, like, build that to make sure that we have the right kind of data set that we are doing and we can sort of generalize across demographics that that are, we have the kind of consistency that it is necessary

in those kinds of predictions. We understand the treatment that's necessary. But the way that we allocate resources in this country often might mean that in deployment, people with Medicare, right, or Medicaid. So that's lower socioeconomic groups or those that have certain social programs may not get that service because it could be reserved

to those that have private insurance. That part of this puzzle in terms of like the insurance of the next step in terms of like how that is deployed isn't just reserved to the person that designed the technology, but also in terms of the way in which we allocate resources in the country, right? So oftentimes those things will go to people in certain social economic groups, to certain communities. We may not get that in an underserved, underprivileged environment. And so that means that that opportunity for someone to get that life-saving benefit may not occur because there still requires this other element of government, this other element in terms of deployment that's necessary. And that's bigger than, you know, what I can do, what we get to do in a technology context, right, in terms of helping people understand the impact or whatever. It's bigger than, you know, even probably what that one, you know, healthcare provider or manufacturer does, right? This is, you know, more about being able to assist the systemically dismantle or

dismantle some of these systemic and institutional things that keep privilege and power. reserved to sell them and not providing benefit to all. And that is probably the thing in this area that keeps me up at night and keeps me thinking about the other kind of connections and the partnerships and the conversations that are still so essential so that we could have this intentionality and development that builds into or gets played out into what happens in the real life in terms of its deployment that can truly provide the benefit that we know the technology is capable of and we're all looking and desiring to see. One other side, Dia, and I know, you know, thank you.

We appreciate your time. I don't want to go too long here. But, you know, what about for the extreme positives, right? So you kind of just mentioned it there through that, you know, example of, you know, AWS customer being able to identify potential, you know, heart issues much sooner, which is huge. But what about just leveling the playing field for, you know, underserved populations, right? Because I think when we think of what happens when AI works, I think so many people, sometimes myself included, just think about jobs, right, and the pitfalls that that could bring and the uncertainty that that could bring.

But what about the bright sides, like leveling the playing field for underserved populations? Right, right. So, I mean, you know, we often talk about examples like in education. And you just mentioned one that I got to go like look at that study, right, like talking about math. Because I remember, I'm old enough to remember when we started using like graphing calculators. school and it was the same sort of conversation about how we wouldn't know how to do math and people become dumber because, right, because we now are using technology. And so that article is interesting, right, to explore, like, what that means because we're hearing some of the same

conversation. But some of the immediate examples that were seeing people leverage, you know, in terms of AI as an education, right, being able to, my mom was a teacher. And so she talked about, like, having 30 plus students in class. And if you had someone that had, have unique needs, right, in terms of how they learned. They, that typically in the class they were optimized to the majority in terms of their learning, their teaching style. And they may not have an opportunity to reach out to one that might have a little bit of a learning difference or we have these mixed classes.

And so someone with that might be a little behind and reading may not get the same attention. So imagine having, you know, AI powered assistance in classes or you have students who English is their second language in the class as well. And now they have power translation to be able to. break and eliminate some of the language barriers, right? All of that now is a way in which AI could level the playing field and ensuring that everyone and all of our learners get the same sort of benefit in a class where typically that might be a little bit more challenging. You know, think about, you know, there are companies that are providing like banking services

to those that are underbanked in regions because they don't have access, right? That's loving the playing field for women and for those that are in marginalized groups that typically may not have banking services like in regions of Manila and in, in, in, in, in, in, in, in, in, in, Africa, you know, assistant tools that, um, are, uh, perhaps not, you know, focusing more on skills when we look at like hiring and can reduce some of the unconscious biases that come into. And there was a company that was out of Latin America that was, like, eliminating some of the elements that typically would be used to, uh, um, exclude and individuals from the hiring process.

That's one of the ways to elevate the, to sort of level the playing field. Think about, you know, the ability to implement systems that, you know, do and provide the kind of robust testing for fairness and bias in systems that are deployed to make sure that social services or financial decisions aren't having the kind of negative impact that we've already seen that they've had for years. Those are the kinds of things that I'm thinking about when we talk about level of the playing field and making adjustments for. And then there are even some simple ways, right?

Like barriers to entry, you know, video production, for instance, used to be very, very challenging and costly, but now folks can do that with tools and technology and enter into a space that typically were reserved to those that had the resources to be able to provide, purchase that equipment. I mean, that is creating and reducing a barrier to entry now that previously existed. And so those are all the things that I think about, ways to bring others along that typically might not have had access or might not have had the

opportunity. But we still have to continue to do the intentional work to have the other processes and systems in place to make that, you know, an ongoing reality. DIA, I think this has been such an educational journey for all of us in, you know, 35 minutes here so far. But, you know, as we wrap up, because we've talked about, you know, the shift in work and upskilling, re-skilling, responsible AI practice. companies should be thinking about how companies and departments should be investing in new areas and goals. But as we wrap up, maybe Dia, what's your one most important takeaway that you want business leaders to remember about what happens when AI works and how we can tackle that with responsible AI? Yeah, I think I mentioned that's in the beginning.

But if I didn't, this is like the kind of key statement for me, at least like one of my grounding principles. I believe that AI can be a bridge, not a barrier. one that, you know, empowers everyone to reach their full potential and to help us drive towards a more equitable world. That does not happen without the kind of intentional action that's necessary to unpack, to understand the areas of risk and to put in place the kind of organizational structure that supports a culture of responsibility and, you know, brings in concepts of value alignment, fairness, inclusion, training and education. the commitment to looking at security and safety, all of these elements to ensure that we ultimately see the kind of benefit that we believe is possible, is possible with the technology. Wow. Y'all, if you heard some click clacking there, that's me typing notes. There's so much, so much good information there from Dia. This is going to be a fun newsletter for me to write.

But Dia, thank you so much for spending your time with us this morning and on the Everyday AI show. We really appreciate your time and your insights. Yeah, and thank you. I know we've been trying to have this conversation for a while, so I'm excited that we were finally able to make things align and get in here together. So thanks for having me. Absolutely.

And we did, y'all. We did mention a lot of different studies and resources and all of those things. Don't worry. You don't got to go chase them down on the internet. They're going to be in today's newsletter. So if we talked about it, don't worry. Just make sure to go to your EverydayAI.com. sign up for that free daily newsletter.

Make sure to read today's, you know, recap. I think it's going to be an important one that all organizations need to take seriously all of these things that Dia was sharing her, you know, her experience in. So thank you all for joining us. We appreciate it. Make sure to join us tomorrow and every day for more, everyday AI. Thanks, y'all. Meet Firefly AI Assistant.

Now live in Adobe Firefly, the Allman One Creative AI Studio. Just describe what you want to create in your own world. and the assistant handles the rest, orchestrating multi-step workflows across Adobe Creative Cloud apps, including Photoshop, Premier Express, and more in one conversational interface. You direct the outcome while the assistant accelerates execution. Stand control with the ability to step in and refine at any time. See it today at firefly.adobie.com.

And that's a wrap for today's edition of Everyday AI. Thanks for joining us. If you enjoyed this episode, please subscribe and leave us a rating. It helps keep us going. For a little more AI magic, visit your everyday AI.com and sign up to our daily newsletter so you don't get left behind. Go break some barriers and we'll see you next time.