Everyday AI Podcast – An AI and ChatGPT Podcast - EP 202: The Holy Grail of AI Mass Adoption - Governance

Episode Date: February 7, 2024

AI Governance is a tricky topic that no one seems to know how to approach. There can be many roadblocks or hurdles before properly implementing governance for AI. Gabriella Kusz, Sr. Fellow at AI2030, joins us to discuss how to create governance for AI mass adoption.  Newsletter: Sign up for our free daily newsletterMore on this Episode: Episode pageJoin the discussion: Ask Jordan and Gabriella questions on AI governanceUpcoming Episodes: Check out the upcoming Everyday AI Livestream lineupWebsite: YourEverydayAI.comEmail The Show: [e-mail] with Jordan on LinkedInTimestamps:01:30 Daily AI news05:00 About Gabriella and AI203008:44 Challenges with AI technology and governance issues.12:02 Rapid pace challenges global technical standards creation.13:26 Ethical AI guidance is available from Fintech.16:41 Draw parallels between ESG and responsible AI adoption.22:33 Industry self-regulation precedes formal laws in societies.26:00 Inclusion and multidisciplinary team crucial for governance.27:57 Clear boundaries for leveraging technology for advantage.33:29 Recognize failures, iterate, and empower AI progress.36:00 Set up task force, admit setbacks, progress.Topics Covered in This Episode:1. Role of corporations and Government in AI Governance2. Organization's Governance Structure3. Risks of AI and Generative AI4. Practical Tips for AI Governance5. Ethical and Global Technical StandardsKeywords:ESG, AI, Artificial intelligence, Emerging areas, Pilot period, Return on investment, Risk identification, Risk mitigation, Governance, Legislation, Change management, Generative AI, Inclusion, Education, ongoing learning, Technology usage, Ethical standards, Global technical standards, Fintech for Good, AI 2030, NIST, Taskforce, Failure in AI implementation, Self-education, AI governance, AI applications, Business growth, AI news.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 and 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. It's that tricky topic that it seems like no one can get a hold of.

Governance, right? There's always these hurdles and these roadblocks that you have to go over or go through until you can properly implement and properly adopt to getting generative AI in your organization. And one of those is governance. So we're going to be tackling that today and more on everyday AI. Thank you for joining us. My name's Jordan Wilson, and I am the host of Everyday AI. If you're new here, thanks for joining us.

This is for you, this daily live stream, daily podcast, daily newsletter. It's all for you. There's always so much happening in the world of generative AI, and it's hard to tackle it all alone. So that's why we are here to help you grow your company and grow your career by understanding and using generative AI. All right, so I'm super excited to talk about governance. It's something we literally cannot talk about.

out enough. But before we do that, I'm going to go over as we do every day, the AI news. So if you maybe are joining us live or you're listening to us on your commute to work, there's always more additional info and links in the show notes. Make sure to check those out, including to our website at your everyday AI.com, where we will be recapping these news stories as well as the show for today. All right. So let's talk about what is new in the AI world news. Well, AI news. the world, right? All right. So a Texas company is to blame for the AI robocall impersonating President Joe Biden. So Texas-based telecon company Life Corporation and its owner Walter Monk were identified as the source behind the AI generated robocalls impersonating President Joe Biden during the

new Hampshire presidential primary, according to reports. So this call was reportedly created with the voice cloning software from AI startup 11 labs, but the company has denied. responsibility. So the FCC and the New Hampshire Attorney General's office have taken action against the source of this fake robocall. So I've been talking about this literally for like nine months that the 2024 presidential election is going to be a lot of people's first foray into generative AI. So yeah, you're going to be seeing a lot of these, probably on a daily basis starting here in a couple of weeks as primary season starts to heat up. All right, next, There is a new and updated AI model for video that was just released.

So Stability AI has announced an upgrade to its image to video latent diffusion model SBD, which promises better motion and consistency in short AI videos. So this model is available for public use now and subscription members can also access it to use it for commercial purposes. So free for education and research, but you obviously have to have a paid account if you're using it for commercial purposes. So the new model is called SVD 1.1, and it's a fine-tuned version of SVD 1.0, and it's optimized to generate, like I said, more consistent in photorealistic AI video. So this space is crowded. You know, I talked about that in the bold takes for 2024, that this is going to be a space to keep an eye on. So not only do you have your runway and your PICA, but now you have the new updated video model from Stability AI, as well as you have Google Lumiere and Meta-Emoo Video as

Well, it's going to get crowded there. And, hey, there's already reports of AI video and Super Bowl ads or big game ads, right? What can you say? All right. Last piece of AI news. The attorney who infamously submitted made up information from an AI chatbot is now undergoing some disciplinary action. So the New York lawyer is facing discipline action for using a fictitious case generated by

AI in a medical malpractice lawsuit. So the lawyer admitted when this happened a couple of months. months ago, admitted to using AI for research and failing to verify the results, submitting a fabricated citation in her legal brief. So the use of AI in legal profession has raised concerns about competency, complacency, and the need for transparency and verification. It's like y'all increase the quality of your input and you're not going to have hallucinations like this. So yeah, this case did happen a couple of months ago, but now the new news here is that it seems like these disciplinary actions are apparently underway.

All right, that was a lot in a very short amount of time. So if something there caught your ear, or we always have a lot more that's going on in the world of AI news and just kind of fresh finds from across the web and recapping our show for today. So as always, you can go to Your EverydayAI.com and sign up for that free daily newsletter. But now let's talk governance, right?

It's something we've talked about here on the show a couple of times, but I think you can't hear it enough. So if you're someone that is in charge of implementing AI within your organization, if you're trying to think of the best ways to, you know, govern generative AI. Today's show is for you. So it's not just me. I'm very excited for our guest today. So please help me welcome to the show.

There we go. All right. So we have Gabriella Kuse, senior fellow at AI 2030. Gabriela, thank you for joining the show. Yeah. Thank you for having me. It's a pleasure to be here today. Absolutely. It's great to have you. And hey, thanks for everyone joining this live, Tara from Nashville and Woozy from Kansas City. Everyone, thank you for joining us.

As always, if you have a question, make sure to get it in now, right? Don't wait. All right, but Gabrielle, tell us a little bit about what you do as a senior fellow at AI 2030. Sure. So my background, I'm a governance subject matter expert. So in international financial sector and economic development programming around the world, 56 different countries. I built governance institutions. I helped to strengthen legal and regulatory frameworks in order to ensure that there was good ethical and standards principles. And now at AI 2030, I'm applying those same concepts to the emerging edge technology fields of artificial intelligence. So working to get the word out, share perspectives through different events, programming, and activities.

like today's and just really looking to help shape standard setting discussions around legal and regulatory and ensuring responsible innovation in the AI space. Now, I'm sure if you are actively trying to implement generative AI in your organization, you are very aware of the concept of governance and why it's needed. But Gabriella, maybe let's hit rewind and zoom out a little bit. But talk about what governance even is for those maybe who are unaware and why it's so important. Sure. So governance usually encompasses kind of two core areas. One deals with, you know, technical standards or practices. And the other really deals with your ethical behaviors or your

moral assumptions for use of a given technology or an emerging area, right? So when we talk about governance, we're talking about the rules or the framework that shapes both the design. Okay. So it goes all the way back and rewinds to some of the engineering days. So it covers some of the design. It also covers the application, so where you can use that technology. And then it also covers any outputs from that technology. And how they're made available to the public or in what ways you can use those particular outputs.

So it's really at each stage, design, build, and then actual use and application. Yeah, and I think it's important to talk about and investigate a little bit deeper because I think sometimes people are under just have bad instruction or, you know, people just think, oh, well, we can just throw a bunch of generative AI into our organization, just like it's a new piece of software. But it's not really like that because, you know, Gabrielle, could you talk a little bit of, you know, kind of the dangers of AI, especially if you don't know what you're doing, you know, obviously

We talked about, you know, the updated piece of news here with the attorney submitting, you know, kind of hallucinations to the courts. But what are some maybe dangers on like, okay, well, here's why you actually need AI governance. Yeah. So I think there's been, you know, a few cases that have come up so far. I believe it was a car dealership that had an AI bot that sold a car for $1 or you have. And, you know, some of these things, they're funny. and then others aren't quite so funny.

So more recently, you have the gentleman who thought he was speaking to a room full of his C-suite board. And in fact, it was, you know, an AI developed board scene and he lost $25 million. So I think, you know, we're starting to see what some of the potential negative implications are. And correctly so, you know, those of us who have dealt with, emerging issue areas or emerging technologies are trying to work together collaboratively to try to both create a framework and an opportunity for stepping forward in a way that positively allows companies, organizations, and individuals to know what is the healthy, constructive way to use this technology and what is not. And I think what you'll see is a need to revamp some of the legal and regulatory

framework around that in a way that's not prohibitive for technology to develop, but also for a way that we start to protect consumers and the general public. So really when we talk about, you know, why governance is important, it comes down to this concept of, well, it's a new space. We know that it's going to be an emerging opportunity, but with opportunities come threats, right? And so that's where we need to start to look at both, you know, some of the security aspects, some of the ethical aspects of using somebody's image without their permission or using images that you've shared publicly, right, through your Facebook account, through your other social media accounts. Who owns that? How can that be used or manipulated? Should it be labeled that it has been

AI generated or AI manipulated? And what implications does that have for society? at large. So that's just a bit of a deep dive using some like very recent somewhat funny, others not so funny examples. Yeah. Yeah. Yeah. And you know, we talked about this on the show yesterday, the example that you brought up the individual who accidentally, you know, transferred $25 million to a company because they were speaking with a very convincing an AI deep fake board, right? Which is something even a year ago, you know, or a year and a half ago, I don't think people envisioned that something like this could be a problem for their company, right? So maybe let's even, before we even dive it a little bit deeper on, you know, governance

and ethical frameworks, maybe let's just talk about, and I'd love to hear your perspective, on just the speed of all of this technology and how that actually complicates governance, because, you know, you can make, you know, the best, you know, rules and regulations, but then there's maybe a brand new technology that didn't exist two months ago when you were putting your framework together. So how can companies really tackle that breakneck speed of new generative AI technologies with being safe yet using Gen AI to stay ahead? Yeah. So I'll talk through some of this at a macro level and then we'll go into sort of the company level. At the macro level, what you have to understand is that this is the speed with which this is moving. So I came from like the blockchain

and digital asset space and that was moving fast. This is just taking off. It's like one day there's, you know, nothing there. And the next day you have a deep fake board that's now extorting or stealing $25 million from me, right? The pace is so fast that it is extremely difficult to build out a framework in rapid pace, right? So just to give you an example, to create either ethical standards or global technical standards, you're talking about a period that usually would take anywhere at the very fastest pace, five years, and at the longer end and probably median seven, and then longer term, 10 years to get one standard through a due process, a standard setting board, committees, advisory groups, ensuring all

stakeholders have had the opportunity to give feedback and input, and that it's not going to have any unintended consequences when it comes to the market, the technology or the geography that it's being applied, all right? Seven, seven years, let's give it on average. Now, if it takes seven years to create a new standard, that creates a lot of bottleneck, and it's also not being relevant or timely for the purposes of users or companies that are trying to have a competitive advantage, okay, while protect their own interests as well as the interests of their customers and clients. So what you're going to see is most likely this gap that exists, and that's where you're

going to have industries sort of coming around and coming together and starting to understand that they themselves need to start to come forward with, at the very minimum, a framework and more likely some level of principles-based approach to standard setting that will allow for companies, especially listed companies that have additional burdens. and responsibilities that they can use this technology responsibly. And so, you know, I think that's one piece of it. When it comes to your everyday company, I strongly encourage people to look at and for guidance from fintech for good

and from AI 2030. We've recently produced a response to NIST, which is our U.S. national standard-setting body, that goes through and provides sort of a high-level etch, around the ethical and appropriate application of AI technology and practice. So, you know, again, you can Google it, FinTech 4 with the number good AI 2030. It should give you some level of direction if you yourself are tasked. If you're, you know, sort of your company is tech ombudsman or you're sort of in charge of

some of the edge technology ethical or standard governance procedures, I would strongly take a look at some of that and also follow NIST. So that's a piece that I think you want to look at as well. And follow what some of our national standard setting boards are doing, the direction they're taking. And that should give you some insight as to like how you yourself maybe need to strengthen order and promote the use of that in a responsible way. So so much good information there.

Yeah, don't worry. This is one of those when I tell people like check the show notes. Check the show notes. Yeah, Gabrielle just just dropped a lot of great information. And yeah, we've had other other guests on the show from AI 2030, FinTech for Good. So we'll put their episodes in there as well. So yeah, geez, so many nuggets there. My gosh.

All right. So Gabriella, let's maybe hit rewind now and talk just about governance because like what you said right there, I think, well, first, we just have to tackle that. Like you can't businesses, like whether you're a small, medium or enterprise, you can't operate the same way you've always operated with these five to seven years going through committees, et cetera. Even with pilot programs, I tell big companies, if you're first. first foray into generative AI as a one-year pilot program, you're going to fail. You've got to have something short and measurable. Anyways, let's talk about just the actual process of governance. Like, what is it? How does it work? You know, because like what you said, Gabriella, like doing the five to seven-year route, you can't really do that. So what's the responsible way to govern

generative AI in your organization? 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 Premiere, 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.adobie.com.

Yeah. So I'll draw some parallels to some of what we're seeing with regards to the adoption and implementation of ESG because I feel like a lot of emerging areas have a lot in common when it comes to having individuals or departments with responsibility for really trying to create order out of what may appear to be somewhat. chaotic, you know, surroundings. And I think that in this instance, you can draw a lot of parallels to that. You're going to have a lot of emerging, likely private sector entities that stand forward and try to create some level of organized framework, okay, or principles. And I think that if you are an

individual who's tasked with helping to ensure the appropriate approach for your company or your division when it comes to AI, the first and best place to look at, I think, will be towards some of like the government, right, and see what's coming out of there. If that's too slow, then start to look at some of the private sector organizations that are trying to advance responsible AI usage, right? The other things you can do is look at some of the larger companies, so the Amazon's, the Googles, the Microsofts, the approaches that they're taking because that can give you some level of insight into how you can build and at least, you know, learn in a similar manner, how to structure an approach that helps to hopefully protect and support your customers and

clients and your staff and organization. I think the last piece I'll say is that you need to be doing things like listening to AI every day and you need to be, you know, subscribing to different news feeds and other, you know, organizational feeds that provide you with insight on how other pilots are being run and structured. I agree with the fact that a year is way too long to kind of test some of this. I think you're looking at more like a, you know, two to three month period, which I think is much more reasonable and also minimizes any of the potential negative impacts that you would have. You need to design that pilot in a way, that you have previously identified what your metrics for measuring success are. You need to bound it

so that it doesn't have leakage into some of the broader activities of your organization in the event that there are negative impacts. And I think you also need to be honest about what the true value is. Return on investing time, effort, energy, and capital. Okay, so what your return on investment is. And then ultimately, understanding and doing a very strong identification of risks and actions that you're going to take to mitigate those risks if you're going to push that towards like a next level pilot. And I would say, you know, again, it's one of these things where you're not going to like do a pilot and then push go. I think you're going to do a bunch of iterative piloting until you and your organization feels comfortable with the framework and structure that you've designed. that appropriately takes into consideration some of the litigation risks, some of the challenges that you're going to see around privacy and protection.

So these are not going to necessarily be things that will be able to be litigated, but it's going to be in the court of public opinion whether or not the way that you've designed and sought to use AI is appropriate socially. Okay. And that's why I say I think there's a lot of parallels to some of what we've seen rollout with ESG, because you're in sort of this gray space of what is and isn't appropriate. And that is ever changing. And so that's why staying abreast of daily news on AI,

understanding what public attitude is towards AI and its applications is gonna be crucially important to designing something that in the end is not only acceptable from a social perspective is legal, what it comes to the shifting legal and regulatory framework, but I think is suitable in terms of the appetite for risk, an appropriate application of that for your own company, its leadership, its shareholders, clients, customers, and staff. Yeah. And I think Gabriella, that it also depends on where, right?

Like a lot of our audiences is here from the U.S., but we have, geez, I think just, I look this week, we have people listening from 160 different countries. So it depends on where you are, right? But like, just let's say for the U.S. because that's where the majority of our, of our audience is from. So let's, you know, you kind of mentioned government. So Raul here has a great question. Raul, thanks for the question. So saying, do you believe that governance will first be implemented in corporations, or do you believe the federal government will have to intervene to make this happen within corporations?

What's your take on that, Gabriela? Sure. So I think it depends on the country. I think if I'm looking at the U.S., you know, if we're also understanding that we have a global audience, I think in the U.S., what you're going to see is due to some of the protracted nature of the legal and regulatory system here in the U.S., the very first steps are going to be corporations that will work to minimize any potential negative impacts to their own bottom line and to their staff and to their products. So it's going to be somewhat self-interested, but the first steps are going to be,

you know, risk identification and litigation in order to ensure that things like the $25 million loss doesn't happen again, right? So that's going to be the first wave of governance that most individuals who are tasked with some role in creating a governance framework for AI are going to be told, this can't happen to us. Make sure you find a way so that this doesn't happen. Yeah. Then I think what you're going to see is an ongoing, which you've already started to see, right? With the utilization of the robocalls for the Biden campaign, for example, now you're

going to have some of the societal pressure, which will, I think, again, before laws are created, okay, there's always some level of industry self-regulation. And whether it's effective or not, you know, that's for the audience to judge. But before there is usually some sort of formal law, the ways that formal laws typically get made is through almost piloting in real life in many cases from industry. And so industry will give input and feedback into that process before a law is, you know, undertaken usually in the U.S. Overseas, you may have, especially depending on the type of government, if it is more centralized,

if it is more a command and control, then you will likely see laws being moved very fast to ensure that some of the top-down powers that be are protected, that those risks are identified and mitigated, so as to ensure no undue or inadvertent unseeding of power. or disruption to the order. But I think when it comes to like more free societies, free markets, you're going to see it happen first in some of the corporations. Yeah. Oh, yeah.

I couldn't agree more. Yeah. Normally I don't interject, but, you know, I'll say at least here in the U.S. because I think people don't have a good picture of this. I don't think there's going to be any meaningful legislation around AI anytime soon, at least here in the U.S. when we're talking about legislation. Will there be executive orders?

Yes. Will there be legislation? Probably not. People don't even realize that there hasn't even been meaningful legislation passed around social media, right? We're still debating the merits of Section 230, which is from 1996. So yeah, I don't think, you know, kind of what Gabriella said, it might be best to tackle this one from the corporate side. I love this question here from Tara. So asking what are the best practices for change management involved in introducing new governance. Yeah, I think people just kind of skip over the fact that those two are very interconnected. Gabriela, what's your thoughts on, you know, introducing kind of AI governance, but then also

still prioritizing proper change management? Sure. So I think in that sense, when you're looking at, so I would say that there's almost like two different levels. One is like, again, I do like a macro and then I do sort of an organizational micro view. At a macro level, I think, we're talking about advancing some of the need for standards and for practices. It's not just with AI. It's with a lot of things for that to be more timely and relevant and to be reviewed and amended more rapidly for obsolescence and applicability. Okay, so that's kind of one piece of that. When you talk about, you know, internally, I always go through a process whereby, you know, you're looking at a multi-state stakeholder group. So it is going to be some sort of units internally that helps to provide

input and feedback into the individual or the department that's tasked with creating that governance structure. So ensuring that you have somebody from marketing, someone from product, somebody from finance, who's part of the conversation around what governance should look like for this space. Because you then have individuals from different departments, those people become a natural delivery channel down into their units and departments. And they've also felt bought into the process. They understand why it's important and applicable to their business function. And so one of the most important things for change management involved in new governance is inclusion. So that's why I think it's really important to make sure that you have that multidisciplinary,

multi-departmental team that talks through what this is as a technology, why we're moving forward with governance and explains, you know, what some of the tangible next steps for people who are ahead of finance need to be. Maybe it's that, you know, when we have important board calls that we also have to, you know, do a text just to double check and make sure that or another mode of communication that's off of that particular Zoom meet so that we now are able to double check some of this. We look through some of the risks both as like the individual who's tasked with the governance setup, but then also ask for feedback from some of those department heads where they see now that they have been introduced to this technology, now that we've started to see what some

of the risks are, what do they see as some of the weaknesses or areas or threats that we need to look at? I think in addition to that, doing some sort of like a basic tutorial that both, you know, educates at the departmental head around AI so that they feel comfortable and confident with using that technology, but also in taking a role in not necessarily calling out, but in shepherding their staff in using that technology and saying these are the things that we're generally seeing are appropriate ways for applying this. Here's where from like an IT or our SISO, you know, talked about. These are the ways that we're not going to use this technology. I just want to make very clear, you know, the bounds of how and in what ways we can leverage this

so that we are not inadvertently exposing ourselves to unnecessary risk, but at the same time that we're remaining competitively advantageous with regards to early adoption and application of what we know to be a very powerful and useful technology. So I think inclusion is part of this. Education is another part, which I just mentioned. And then I think lastly, because this is an emerging space, it's ongoing education.

So that little task force that you'll develop, that's multi-disciplinary, multi-departmental. It is not going to be like a one and done. Like, glad we had this talk, guys. Let's never waste people's time again. It's kind of like a, hey, once a month, you know, I'm the head of IT or I'm the person who's been tasked with governance. My job is to listen to, learn from, like, you know, again, with emerging technologies,

YouTube videos, podcasts like this one. And then, you know, to be responsible for really kind of giving people, I don't know, do like a, you know, brown bag lunch session once a month so that people don't feel like it's like they're being forced to do more work. I mean, I always pay people off in cupcakes and cookies, so, you know, that's kind of an easy one. And then just say, like, this is what we're seeing on the horizon. These are the key trends. I want to make sure that people are aware of what these things are, you know, and just keep people up to speed on some of this. Encourage, especially younger people in your staff who you're prepping for succession planning and future

leadership, that they're aware of this, that they're keeping pace with it, and that you also have somebody else who's kind of out there watching, especially as it relates to their particular vertical. Yeah, something you said in there, Gabriela, that I love just the risk mitigation. And it seems like sometimes people are so risk adverse with generative AI, which makes sense. But then it just leads to inaction, right? And then it leads to, you know, losing ground on your competitors. So yeah, you have to balance risk mitigation with proper implementation, I think, It's huge. Great question here from Monica. Thanks. Thanks for this one, Monica. So asking what are some of the mistakes that maybe you've seen that we can learn from, Gabriela? Yeah. Well, again, we'll do some of

those that are publicly available and have the consumption just to, you know, be sensitive to some of the firms that have, you know, forayed into this space and have had some kind of negative experiences so far. But I think one thing is sort of like a blind idealism towards what this can and should do. I think just like when the internet was first coming forward, there's going to be a lot of ways that this can be applied. Right. So it always has to be like a cost benefit of like whether it actually makes sense for your department, for a particular product, for its particular customer segment, right? So I think that one of the things is just like, yay, we have a new technology. solves everything. Now, I expect you to get your work done in 10 minutes instead of 10 hours,

right? So I think, you know, some of those pieces are going to be important. So blind adherence to or excitement around a technology in its application. Mistake number one. I think mistake number two is like the exact opposite of that, which is a complete and utter clamp down, which is that we don't know what this is. No one should use it at all. It should be. So it's almost like I think the main mistake here is kind of around balancing it, being a learning organization that sees the value, knows how to dip their toes in, and is doing it to the best degree possible with regards to responsibility. I think you want people to test and play with this technology, and it shouldn't just be at some of the higher levels. You want it to be encouraged with younger staff. Why? Well,

most likely the younger staff is going to be more open to using it. They're going to be the ones for whom this will dictate sort of their trajectory professionally and success. They're also going to see it as an early indicator of the degree to which your firm and organization is open to transformation, opportunity, and kind of the next wave of where your product, market, customer client is going. So I think that, you know, if I'm looking at some of this, it's, you know, one, blind adherence, big mistake. Two, ultimate clamp down. Second mistake. Three, though, is keeping it almost in like an ivory tower, right?

And saying only this high level group should be able to apply it because probably the better applications of it and the people who are going to be more savvy around it are going to maybe be some of your like newer hires or your like lower level management. And so I think that making sure that those people feel both empowered as well as supported. And then lastly, I think, you know, if I had to name a fourth issue that I think comes up, it's that like all mistakes are, you know, irreversible and no one should be forgiven. It's a new space. There's going to be lots of mistakes that are made. And if the reaction from senior leadership is a complete, like you're fired, you're to do this, it failed, or like you were saying, around piloting.

There's going to be lots of pilots that will fail. Acknowledge it's a failure. Don't just stop, you know, iterating or playing with what could be potential applications of AI, but really start to think about what is it that would have been better, you know, and really do a true debrief from that, right? Because a lot of times we're like, oh, AI didn't work. So the entire technology can be thrown out. And you're like, well, hold on.

Like maybe it didn't work in finance, but it definitely could work in, you know, engaging around business development and creating prompts and scripts and, you know, more personalized tailored emails, saving time, you know, efficiency effectiveness. You know, what are some low risk areas that we can apply this first that don't unnecessarily, you know, disclose proprietary or sensitive information, but can enhance people's daily job experience? so that they're able to be freed up to do some of those higher value added tasks, right?

And I think that's where you're going to find those like kind of like lower management, maybe even higher level, early higher people who I think, you know, they're going to have some really cool ways to work with this. And I think it's really important to empower them and to support them in sort of a lot of that inventing. Yeah. So so many, I mean this episode, Gabriela has been like a good. a goody bag of governance. You know, we've talked about how to set up ethical frameworks, best practices on working

with stakeholders, balancing risk mitigation and timely implementation and, you know, inclusion, you're right, involving all levels of an organization. But, you know, as we wrap up here, maybe what's your one best piece of advice for someone, maybe they are in charge of, you know, governance or they're in charge of implementation? What is that one very specific practical piece of advice that you can give for people that they can start implementing AI responsibly with governance in mind. Yeah. So I think I'll give you two and I'll make them really fast because I know we're getting close to wrap up. But I think the first one is to make sure that you're learning. And it's not going to be your traditional learning by the time

formal educational resources are available. You know, they're going to be so watered down in general. you really do need to start to do your homework by listening to podcasts, by going on to the, you know, onto YouTube, watching videos. You need to self-educate on this so that you're ahead of the curve. That's number one. And number two is get a task force set up immediately. You know, start pulling people to the center, start talking about what this is. And don't be afraid to admit when things fail. You know, you learn so much more from when a, you know, pilot fails, fails. And I actually don't believe in the concept of failure, to be quite honest with you. I think everything is just a part of the ultimate success story. So, you know, there's stepbacks, but I don't know that you

could ever consider a failure or failure if you learn something from it. So that's why I think it's really important to see that as a progression towards a successful application of this. Jeez, this has been a lesson in a half. Thank you. Gabriella Kuz from the senior fellow AI 2030. Thank you so much for joining the Everyday AI show to talk governance. We appreciate your time. Thank you. This has been fine. Hey, and as a reminder, everyone, this was a lot.

Make sure to check out today's newsletter. So go to Your EverydayAI.com. We're going to be recapping that. Tomorrow, join us for our show, Translation in the world of AI. Will we even have a job tomorrow? Tune in and find out. But thank you so much for joining us. Hey, you know what?

I'm going to go ahead and shout another episode. So if this was up your alley, if this was super helpful, what you heard Gabriela talking about, go check out episode 197, which was our five simple steps to using Gen AI at your business today. These two episodes work well together. So thank you for joining us,

and we hope to see you back 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 words

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 EverydayAI.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.