Latest / The AI Governance Collective Podcast / Episode 0: Meet Your Hosts
Transcript
- Dr. David: Welcome to the AI Governance Collective Podcast. My name is Dr. Kyle David. Emikly Dorlack: Hi everyone, my name is Emily Dorlac. Dr. David: Hey Emily, how's it going? Happy Halloween Eve. Emikly Dorlack: Right. Halloween Eve. I'm I'm already spooked about how this is going to go, but I think it's gonna go go well. You're all here. And yes, Halloween. Thankfully it's on a Saturday. for those of you out there, parents or not, you know, maybe your neighbors of young ones, having it on a Saturday, so much better than a weekday, right? Dr. David: much better. I think the kids prefer to have it on a weekend because then there's not an excuse to go home at nine thirty or ten. they they can argue to stay up until midnight, or at least that's what my teenagers try and do. So Emikly Dorlack: Yep. I'm sure as an educator you can also appreciate that. I can't imagine being those teachers the next day I'm always like, Thank you for what you do. You don't have to give them back to me after school. I'm kidding. I'm kidding. Dr. David: All right. No. No, that's good. All right. So this is the soft launch of our new podcast, the AI Governance Collective Podcast. we're gonna have more for you on that soon. But what we'd like to do today, a combination of two objectives. So, first is just for those of you that are that are new to the collective, new to Dr. David and and to Emily, we just wanna introduce ourselves. We wanna let you know a little bit about us, how we came to AI and Why we think that having this type of podcast is important. We're also going to talk about the objective of the podcast. What is our vision for this and for the community? How we want to get you involved. So and then after that, we'll take your questions. I see we already have some questions coming up in chat. So as you're joining us, if you have any questions about IEPP exam prep, about the AI governance world, career coaching, we're we're happy to jump into all of that. So so Emily, maybe we'll start with you. Tell us a little bit about yourself and how you came to AI. Emikly Dorlack: Sure. Thank you so much, Kyle. Well, I think it's probably important, especially with Veterans Day being upon us, and it's also timely and relevant. My formative years before my career of were spent being part of a military family. My dad was in the United States Air Force for 37 years doing special ops. And I say this because I've really started to realize later in my in my life that service, strategy, and community, they're just part of my DNA. but more or less come from a military family. My dad really instilled strategy and operations. I always have a plan A, B, C, and D. people encouraged me when I was younger to become a lawyer. Didn't know what that was. I love to read, I love to write, I overanalyze everything, but I'm in recovery for the most part for that. And I studied public affairs and environmental affairs at IU, Root Root Hoosiers. In case you're wondering, we did win the national football championship. I studied public affairs and I loved it. Thought I'd work at a think tank. More or less my career path took me to law school. I did work at some firms in my formative years professionally. And then when I was in law school, I worked for the government. I worked for Cook County in civil and in criminal proceedings. Thought maybe I'd be a prosecutor. And really felt like I was going to break the norm. I did not want to work at a firm. I certainly did not want to work in the court system either, but I had a really big appetite for finding a way to merge technology, public affairs, and also just strategy, right? And using that in a way that adds value to society. I ended up in healthcare and I was at a major health plan. I worked in procurement and that was a big cross-functional role. I worked for the digital sourcing team in the latter part of my career there. And I am the product of Wonderful Leaders team. I must say that. Like this really is the product of people that saw potential in me or people that tried to find ways to capitalize on my boredom at work. And one day I was in the middle of a discussion and I saw the words responsible AI program. And I thought to myself, whoa, what is this? What is this? truth be told, that night I didn't sleep much. I went down this rabbit hole of research to learn everything that I possibly could. this was also in 2022 team, right? So there wasn't much out there. Yes, I stumbled upon IAPP. And the next day I I woke up and literally said, I think I'll be an AI governance advisor today. Just like that. I reached out to one of the leaders and she was fantastic. And so supportive, and I more or less was determined that I was going to make this work and that this was for me. It it brought together so many disciplines that I've been working through my entire career. Technology, law, policy, procurement, vendor relationships, organizational structuring, cross-functional relationship, and working with complex language and making it simplified. So I pivoted. actually within two weeks because there was an opening on the team and I I I wanted that role. It was a rigorous interview process, but I got it and I found myself. This is where I belonged. I've completed just under 300 assessments in the vendor and the impact assessment space. And I don't say that to brag. I just say that to mean I know what it's like to be in this field. I know what it's like from an organizational perspective to have lots of things changing. And also to try to keep up with learning the technology, learning how it fits into your own organization. And that's something that I'm really passionate about. So, roll back a year and a half ago, I had an opportunity to get out of healthcare space, was there about nine years, and I always call it the highest level of scrutiny, right? If you can go from running a full-scale responsible AI program in the healthcare space, it's good to test the waters elsewhere and an opportunity. found me actually through my own reaching out. There's a theme here. There's a theme here, team. I have this itch to do something and I try to make it happen. I had seen a job posting and it looked like it fit my character and what I was looking for and what I could do. I got dismissed right away. Now, mind you, I'm fresh off of several impact assessments for core and non-core functions. And I said, why'd I get rejected? Help me understand this. Not thinking that I was going to get a response. And they said, actually, what really happened here is the person decided they don't want to come back. The person was going to come back. It's open. You're right. What's let's run this by different recruiters. And I had some interviews and they said, You're right, you're a great fit. I'm I'm very humbled by that. So I got to work on a consultant team for for Microsoft and got to see AI governance at scale there and another leadership moment. A leader that I worked with closely during that time a couple of months ago, actually, Kyle, right before the AI Governance Collective started, randomly told me, there's so much more in you. What do you want to do? And this really stuck with me, right? I had no idea what this meant. Am I not performing well? Am I in the wrong field? And what I came to realize is that I had a lot of experience that people could benefit from. And I was missing that community aspect of my work. And I'm not to not to say I don't know everything. I don't have all the answers, but I wanna find them out. That's that's a part of my DNA. And lo and behold, the AI governance collective came into my world and I have not I have not looked back. I I love it. It's been a wonderful experience. And Kyle, thank you so much. I'm I'm humbled and flattered that you chose me to do this. And also on on a side note too, there's been this running joke with some circles of my friends professionally and personally and they said, why don't you have a podcast? And I and I've told them there there are many reasons why I have a podcast. Do you do you really want to give me a microphone? Kyle, we did not run an assessment on this. Like I think we should document it. because I have limitations myself. And I I'm thrilled. I'm glad you're all here. We hope this is a a really valuable space. And Kyle, this is really a huge tribute to you. Thank you for seeing this need in the world. Dr. David: Well the Thank you for doing this, right? I I'm I'm the one that should be humbled and and grateful that that you decided to join the team. so yeah, glad you're here. And w we started a podcast. Our our first goal is is to is to finish this one, right? but we do intend to make this weekly. We have a lot of really exciting interviews with a diverse group of experts in this space that we're really excited to bring to you in in the coming weeks and months and and over the next year. So yeah, there were a couple of things that you said that that really spoke to me. so the the pivoting into AI governance, right? You you saw that term, you were attracted to it, you were determined to get involved. I think one thing that I really enjoy about the privacy community, and I found to be the same for the AI governance community is our backgrounds. you don't get a degree. I mean, maybe you can get a degree in in privacy and data protection. Nowadays. But I think that's going to be a relatively new offering. Similar for AI governance as well. You know, there are a lot of certificates that are out there, but I don't think you get a degree in AI governance and then you get an internship in AI governance. The people that are in the field doing the work now, this is something that they maybe fell into, or it's something that sounded interesting. It was an opportunity that, you know, something that they wanted to learn more about. And And they took a little bit of initiative. They asked some questions, maybe sent cold emails, did follow-ups, and they got involved that way. And I know we've got a lot of people that are interested in pivoting into this space as well. The main takeaway for me, and I think a takeaway for the audience, is that there are spaces for everyone here. You don't need to be a technical expert. You don't need to have created a large language model from scratch or understand all the technical controls and how to implement them. them that is useful background. And I think learning on the job is an important trait for anyone that wants to be successful. But you don't have to have that specific background in in order to get your foot in the door. And then another thing, I can't even read my own handwriting. Emikly Dorlack: We have that in common too, Kyle. You should see my handwriting. Or or not, you won't be able to read it. Dr. David: and then like the the public service aspect of you know how that how that called to you, you know, wanting wanting to serve, wanting to to give back to to communities is something that resonated with me as well. My I I'm a lifelong educator, or at least for as long as I've I've been an adult. going on 20 years now, I've taught first English overseas as a foreign language, and then Chinese and world history at UC Irvine while I was completing my doctorate at the Department of Energy here in DC, privacy, data protection, specifically federal privacy. And then when AI started to take off, they needed an AI literacy program, and I helped to stand that up as well. Well, and now IAPP certification and so forth. But my interest in education, why I wanted to get into that, was it it wasn't because I was a good student, I was an atrocious student. And it wasn't until I was a young adult that I realized the problem wasn't with me, the problem was that the one-size-fits-all system just didn't work with the way that I learned. And I wanted to be an educator so that I could help other people like me that struggled. initially I thought, you know, university was gonna be a good place for that, but you know, I I am where I am now. I I have no regrets. I'm very happy with I'm very privileged to be in a position where where there are people that that want to learn with me and enjoy the content. And and so pivoting a little bit into our our vision for the podcast, that's one thing that's important. For me, is the implications, or you know, what is happening with artificial intelligence, its tremendous transformative potential, as well as its risks, are talked about all the time. That's all my LinkedIn feed is filled with, and you know, the dozens of newsletters that I subscribe to as well. very little of that. Content actually tells us what to do about this, though. And for me, AI governance is, you know, part of it is about trust with the public. you know, using privacy as an example and and how responsible data protection is important for trust in institutions, trust between you know, companies and and the public and their customers. I I see AI governance going the same way. And, you know, if if we consider, you know, AI has been around for decades now, but if if we're thinking about generative AI specifically, then, you know, this moment is only a few years old. And all of the polls tell us that the American public at least is very distrustful of AI. And I think they have a lot of good reason to. One, Of my visions for this podcast, one of my objectives is that we can help AI governance professionals, AI risk management professionals not only help their organizations appropriately mitigate and control for risk, but also help build trust with the public at large. That we can minimize, you know, with privacy these The breaches that we've saw, where hundreds of millions of people lose their data. I I I don't want AI to go the same way as privacy. And there's only so much we can control there, but I would like these professionals to feel better equipped for their jobs to help their organizations and and to help repair some of the risk that we're seeing with or reputation of of AI, so to speak. it's a transformative technology and and I think if we deploy it responsibly, it's it's gonna have a tremendous impact on us as a society. what about you? What are what what are your thoughts? What are your goals for for the podcast? Emikly Dorlack: Or something that I didn't recognize. was lacking as well until you said, Hey, there's this awesome, cool thing I'm gonna do. What do you think of it? And I said, Where's this been the last four years? it's definitely hard to be in a space that's constantly changing. It's equally hard to be in a space where you're highly regulated for every single thing that you do at your organization. And there's this inherent duty in my mind to everyone, everyone in society to understand this technology. I think people are doing this work, they're navigating ambiguity. Whether they're organizational constraints, maybe it's competing priorities, maybe it's incomplete information. I think people need a space where they can go and compare experiences or even problems and and learn from each other. I can say this too as a deep procurement. individual with lots of platform and SaaS experience and exposure contractually and just operationally. The AI governance collective, it is not just a platform to me. And I want everyone to realize that this is supposed to be an opportunity to help develop a community for for educational resources, to to go through this together. Let's have these conversations. knowledge so knowledge sharing definitely knowledge sharing practitioner development because we have so many personas and I think some of you that have I've met or spoken with there's a place for everyone in AI governance maybe not the same level but there is a place I also think constructive disagreement and professional connection is is vital to the success of society And all of us professionally, and I think we can translate theory and frameworks into practice. There's no manual for the how. It's tough. It's tough. It's definitely inspiring. It's it's definitely complex. It doesn't keep me bored at all, but it's hard. And I think that the more we're open to having those conversations, the more we can bring people together and benefit. how we look holistically, whether nationally or internationally, at AI governance, because AI is not going anywhere. And we're going to adapt from a governance perspective or we will get left behind. Pretty much. Yeah. Yeah. Dr. David: Yeah, yeah, I agree. two two things just to piggyback on there. You y this was this was implicit. I just want to call it out. Psychological safety. That members Emikly Dorlack: Yes. Dr. David: members of our community can can come to us with questions and not feel embarrassed or uninformed. We're building the plane as we fly it. and so building a you know a virtual space that's available 365 days a year, I think is is really important. one thing I've really liked about all of our workshops is we're getting a tremendous amount of audience participation. We have so many questions that we can't address all of them in, you know, an hour long workshop, for example. But because we're we have an online platform, we can take those questions, we can put them, post them in a related space or in the event space. And we can continue Emikly Dorlack: Right. Dr. David: the conversation afterwards. And then all of that is is captured for posterity, right? For people that watch the recording a week or two later. and you know, I I recently went to IAPP's PSR. it was in Seattle a couple of weeks ago. It was it was the best conference I've ever been to. I I really like that conference. It was the my first time going to PSR. I recommend it for everyone. I left energized. I also left with a spiral notebook full of really fantastic insights that Emikly Dorlack: Mm-hmm. Dr. David: I I haven't yet acted on, that are losing context. And I had a lot of really good conversations that died in the hallway. And so one vision Emikly Dorlack: Mm-hmm. Dr. David: for the collective, something that I I would like to, you know, continue or or inform our podcast, are all of these discussions, right? Actually having Emikly Dorlack: Sure. Dr. David: a central online virtual space where These conversations, these insights, these resources are are curated and preserved so that they become a a resource that continues to become enriched throughout throughout time. Yeah. Emikly Dorlack: I agree with that too, Kyle. And coming from the view of someone who's asked if they're interested in putting on a workshop series and and what am I gonna say? What can I teach? How can I make my experience industry agnostic? framework agnostic, but helpful, but real and tangible. I think what I've learned about this community as well is everything doesn't have to be a formal, perfectly packaged presentation. I I don't have all the answers, Kyle. I know I know you don't have all the answers either. I think one of the most valuable things that we're developing is a collective pun absolutely intended collective body of judgment of strategic thinking and of safety to acknowledge here are gaps whether it's at your organization whether it's in your own personal use with AI and what can we learn and where can we go from here? Dr. David: Yeah. All right. you want to get to some questions, Emily? Emikly Dorlack: I have one question for you before we dive in. Uh-oh, okay. Dr. David: okay. Emikly Dorlack: Every superhero has an origin story. And I think at what point did you realize you wanted to build something like the collective? And which superhero do you think you identified with at that time? Dr. David: boy. So the the the collective answer was The inspiration from the collective came from a couple different places. so the first one is So I've been, you know, providing training for the AI governance professional certification exam for for over two years. I was one of the first people to get certified right after the exam was launched in April of 2024. and over the last going on two and a half years now, the number one question I get from folks that pass the exam is now what? What do I do? Emikly Dorlack: Sure. Dr. David: where where do I go to learn more? Where do I go to get hands-on experience? And I wanted to create a place, a a a virtual community or really Emikly Dorlack: Mm-hmm. Dr. David: a hybrid community, because we have two in-person local chapters now where where folks could do just that. And, you know, we have organizations, global organizations that have people involved in AI governance, they have events, but there's nothing in between the events. It you know, my comment about the conference I attended a couple of weeks ago, really great, gave me a lot of energy, learned a lot. It was two days, Emikly Dorlack: Sure. Dr. David: and then it was done, right? and so creating a space where you know that knowledge, that community is there 365 days a year. So it was it was about you know, seeing that there's a a community of people that want to do more, that want To continue their education forward, that want to network and learn from others, people that have common struggles and can very clearly learn from each other. One other thing you you said earlier with regards to knowledge sharing, in my experience, and I talk to a lot of people doing this work. Folks are often focusing on one part of the full AI development lifecycle. They may be involved with use case assessments or strategy and policy. They may be doing AI incident response. Emikly Dorlack: No. Dr. David: You know, they may be doing any of a number of other activities. Very few people are doing full life cycle management from use case assessment to exploratory data analysis to evaluating performance. Performance metrics on a newly trained model, to ongoing monitoring and maintenance, to decommissioning, right? and everything Emikly Dorlack: Shim. Dr. David: that's involved in that. And I think the the best AI governance professionals in the future may still just be focusing on a handful of those stages, but they will Emikly Dorlack: Mm-hmm. Dr. David: understand and they will they will have a very good grasp of the full life cycle because that's going to allow them to understand everything in context. And that's another thing I wanted to do with the collective was give folks an opportunity where, all right, maybe you're just doing, monitoring and maintenance, or you're just doing AI instant response. But if you want to learn more about the nitty gritty gritty of data collection and preparation and privacy enhancing technologies, of what it actually looks like, what the code looks like to train a model, what type of metrics, you know, red teaming and so forth, I want, I want there to be a place where those resources and that education. education is available. so now the superhero question. gee. Emikly Dorlack: You don't like superheroes or comics, so I thought it seemed appropriate. I I know you don't like Lord of the Rings. Dr. David: Uh-huh. Yes. Yes. Yeah. I I I am like a big fantasy person. Like the best fantasy series ever is the Wheel of Time series. Amazon completely butchered it. That's why it was canned. I I have a lot to say about that. Emikly Dorlack: Okay. Dr. David: I I I don't know. Like I think the educator in me would probably appeal to like so which one of the X Men is it is it the beast? Or a Bruce Banner type, right? educator, community builder. I I don't have much of a dark side, I don't think. Emikly Dorlack: No, I don't never sense that at all. Dr. David: y y yeah. So I I don't know, like so like Bruce Banner, but not the Hulk. Is that allowed? Emikly Dorlack: It's a hybrid. Of course it's allowed. All right. Dr. David: Sure, sure. Okay. All right. all right, let's let's jump into some questions. Emikly Dorlack: Sure. Dr. David: All right, so we've got one question here from Ashvini asking about whether or not the exam, the AIGP exam results, or or are you graded on a bell curve? To to the best of my ability, I don't think there is a bell curve, but IAPP hasn't shared a whole lot about how they do their grading. what's available is in the general candidate handbook. and then you have the weighting that's available in the the bodies of knowledge. you know, how many questions, like a minimum and maximum, how many questions can you expect for each of the domains and from the different competencies? but whether or not they're they're grading on a bell curve, I I haven't seen anything that speaks to that. Emikly Dorlack: I certainly hope not. For what it's worth, I have a lot of thoughts on that that I won't bore you with now. Dr. David: all right, let me see. So another question here from Raphael. He's asking, really this is an experience building question. He's asking specifically about like voluntary projects to take on to build experience. it doesn't I'm assuming probably AI governance-related experience. So Emily, do you wanna start with that one first? Emikly Dorlack: Yes. wanna frame this appropriately. Wherever you're based, it's very possible in an educational view or community view, whether it's church, there are conversations in those spaces about governance. And I don't necessarily mean they're bringing in Jim Wiggins or something like that, but but they're people, they're real people just like you, just like me, that are trying to learn about this space. And sometimes I'll see an ask in the community. From a public perspective, hey, we're getting together this group or this cohort or this chapter, or we want to speak around this or this. Can you provide some insights? I'm normally one to turn that stuff away. I'm not a big self-promoter, but I'm very big on community building and knowledge sharing. And I highly encourage you to get involved in your local community. I also would encourage you if you have any contacts in the public space or in the corporate space to have. Coffee chats with them, catch-ups to see what their experience with AI has been. And you might be surprised about challenges that they're facing or opportunities to help, perhaps on a ten ninety-nine capacity. Who knows? Who knows? I think that's one thing I've seen, Kyle, is there's been a shift where people freaked out. about what governance they did or didn't have. Now they're starting to operationalize it. And now I think some people might be afraid to say we're still figuring it out. What do we do? Because there's this expectation that we have it figured out, no matter where we are. And and that's certainly not always true. Dr. David: Yeah. w one one thing I like about our little technical mishaps that have have been occurring throughout the episode is It it it's humanizing us, right? Emikly Dorlack: Yes. Dr. David: and that's you know, one thing about the the collective I think that's important about psychological safety. People can come with with questions and you know, share what they're going through, you know, with their organization and and feel like they are not alone, that there are other people that are going through very similar situations. no one has this figured out, even folks like us. us that have been doing this, over over the last three going on four years now. and again, you know, someone may be Emikly Dorlack: Yeah. Dr. David: very well versed, you know, have really good expertise in in one area, but be be weak in another. So I I I agree with that. Some places to look if if you're looking to build experience in AI governance. So the very first place would be your current organization. If you're currently employed, looking to see what functions exist within your organization, if you're a student, taking a look to see what's available at your school. And if there's already an AI governance function, if there's a GRC function, reaching out to those people. And at the very Least, maybe like Emily just said, getting a coffee or or going out to to grab lunch, just to talk about, you know, hey, what type of work are you doing? How did you get involved? Do you have any openings expressing your interest in pursuing this as a career and just learning from them as much as you can? So whether it's an employment opportunity or a learning or a volunteer opportunity, there's there's something to learn from those individuals and the networking component of that. That is very important. If you find that your organization does not have an AI governance or a GRC function, then maybe that's something that you can volunteer for. again, I Emikly Dorlack: Good point. Dr. David: think I I can't. I'm sure there are organizations, there are businesses out there that are not considering artificial intelligence. I mean, I find that hard to believe, but that's just because I'm immersed in this every day. I I think most of you are going to find that either. employees, colleagues, other students are if they are not using it professionally, they're thinking about how to do it and helping people understand, you know, just providing a very basic AI literacy foundation, understanding risk, understanding how the technology works, its benefits, its limitations, that's that's one way to get in there. So first point, current organization do they have a function for AI governance? Do they want a function? And even if there's not room to build out that function, maybe you could do something a step down would which would be like a community of practice. So if you're not getting buy in or a a mandate, if there's not a budget line for an AI governance professional, could you at least do a community of practice? What can we how can we deploy or how you know? What can we do with AI in the organization to make us more efficient, to help us do more with less? the second would be looking for nonprofits or NGOs in your local area to, you know, they if if it's a nonprofit or for or if it's an NGO, they may not have many resources to support a full-time person, or maybe even like someone on a fractional basis. But if if I was gonna start a consultancy. on my own today, the very first thing I would do is I would actually try and do a couple implementations for free to build the experience to get the testimonials. And then that way when I went to charge for the service, then I could say hey, I've done this for a handful of organizations. Here are some testimonials. I did NIST AI RMF with this and, help this organization you know put together compliance for the EU AI Act and and and so forth. Emikly Dorlack: Sure. Dr. David: so nonprofits, that's another one. Last thing for trying to build experience outside of getting hired, look for the large global professional organizations like IAPP Emikly Dorlack: Mm-hmm. Dr. David: and Isaka. They have chapters all over the world. if you don't have one local, look for online organizations like the Cloud Security Alliance or OWASP. there's another one that escapes my mind right now. But the the Cloud Security Alliance is always taking volunteers for all types of of really interesting projects and white papers. OWASP does as well. And other organizations like women in AI governance, these are really good Emikly Dorlack: Good job. Dr. David: online communities to to get involved with to see, you know, what type of hands-on experience there is. and then of course, this is what we're trying to do with the collective as well, is through through workshops, through, through training. build out opportunities for people to take these frameworks and and apply them to to start Emikly Dorlack: Okay. Dr. David: building that skill set. Emikly Dorlack: One more thing you said there that I wanted to just hit home or add a little bit more flavor to as well is One of the most insightful pieces of my journey has also been talking to current leaders in AI governance, different personas along the life cycle, and hearing where they came from. It's really, really intriguing because you can have backgrounds that translate very well with cross-functional disciplines. And we all have to learn. It's a transition for everyone no matter where you come from, and that's okay. Dr. David: Great. Yeah, th there are everyone has transferable skills. It's just a matter of learning how to frame those in a way that very clearly Emikly Dorlack: Right, May? Dr. David: demonstrate the the value that you bring in organization based on that on that background. Emikly Dorlack: Great question. Dr. David: All right. Very good. Let's see. all right. So we've got a question here from Michelle about so it looks like about the the NIST AI RMF templates that we're creating. So so within the collective This would have been, I think, in September we stood up a NIST AI risk management framework working group. we had two reading sessions where together as a group we we read through the core RMF. We had an eight-hour workshop with Jim Wiggins last Saturday where we methodically went through again the the core RMF. And then not next week, but the week after, we're starting reading sessions on the generative AI profile as well. So again, in in wanting to provide education and and training, you know, these have been volunteer led efforts and they've been very well attended and and we've had some great discussion. Our primary output For this working group is going to be a library of 27 templates, all aligned to the NIST AI risk management framework. The objective is that the library will allow anyone to assess, stand up, and mature an AI governance program that's based on NIST. So we we had a couple forms. Well, we had a form go out over the last week or so, just having people volunteer to lead on different artifacts. And I'm going to be publishing that today, as well as the first three artifacts that we're going to work on. I'm also going to be publishing the the volunteer assignments. So every artifact will have two leads as well as two reviewers. And these will also be open for the community to go through and add their comments. the comment period will be open for one week, and then for that second week of the sprint, the the leads will be responsible for harmonizing the comments, addressing, any concerns or blockers. Once the leads are finished, they give it to a QA team. the QA team makes sure everything looks good, and then they're gonna send the templates to to Emily and I for final review and sign-off. As soon as those templates are signed off on, we will publish them. They'll become available right away. So those 27 templates, the their release will be staggered over the next four months or so. So we start early November and we should have them all published by early to mid-March. So November, December, January, February. But we're basically taking a month off for the holidays. So over over three months. but those assignments I plan on publishing them as well as a short just demonstration video, you know, how to work through the spaces, where to find the links and and comments and so forth Emikly Dorlack: Sure. Dr. David: later today. So if I if I don't get to it today, it'll be tomorrow. But all that will be up by Monday for sure. Emikly Dorlack: Can I also just say really quickly, briefly, I love those resources. As someone that came from this non-AI collective world, I have been dense in frameworks for so many years, HIPAA, you name it. They're dense, they're hard to understand, it takes time. And I love that there's this space for people to be able to go through it and understand the policy embedded within all of that. And also just to see how terms are used and to draw patterns because if you don't understand, at a moderate level minimum vScore frameworks, it's going to be very challenging in any part of the AI lifecycle to do governance. Dr. David: Yeah. Yeah. and I think you know the the best way to learn these frameworks is I I guess there are three components to it. So first to to read through it. And the the NIST AI RMF is very approachable. It's in plain language. It is not very technical. When there are technical terms, I feel that they're explained in again in plain English. the Emikly Dorlack: grade. Dr. David: second would be to actually have a template, like a a template library to read through. So you can see the types of questions that are being asked or the the information that you're gathering in in order to to perform your your risk assessment or to put together your risk Emikly Dorlack: Mm-hmm. Dr. David: tolerance statement and so forth. So actually seeing the instrument that you would be using as you're implementing the framework. And then the third component would be to actually Actually apply it to Emikly Dorlack: for. Dr. David: you know, even if it's a fictional scenario, to kind of do something like a like a murder mystery where you're given you know different fragments or clues, and then based on that, you you fill out the template. So each of our templates has three parts to it. There's the template itself, which is blank and ready to go. The second part is an append, or there's one appendix which provides guidance on how to fill out each of the sections, you know, why. This section is here. Everyone ties very explicitly to one of the 72 subcategories throughout the RMF. And then the third component is a worked example. So again, the idea here is to produce a resource that members off the shelf are able to use and start implementing. and that's actually a really good segue to a question that Chaya has asked here. So it I I believe Chaya's question here is she's she's reached out, she's she's tried to make the case for AI governance and and the value that a governance program or even something like a a fractional chief AI officer or or something could bring to an organization. And it it seems like one of the responses she got is all you're going to give us is a form that gets filed away, something like a like a SOC two certificate. So Emily, do you want to take a stab at that one? I you know, if if you were if you were pitching AI governance as a as a service, right? As you know risk management, what what would you say How could you get buy-in from a company to make space for that and and to cooperate for that type of risk assessment? Emikly Dorlack: Sure. It's possible too, I'm already thinking of merchandise potentially for the AI Governance Collective and my mug would be, can you please walk me through that? Can you unpack that reasoning? So first, I definitely wanna understand their why, what's their basis for suggesting that this is just a form? It is not just a form. Secondly, I think as a part of understanding where they're coming from, are they aware of the usage of AI across the organization or institution? Are they aware of any information that is given to those employees about AI literacy concepts, AI usage in general, or is it a free-for-all and people can do what they want? How close is the organization to a regulated sector? What's the industry we're talking about? That's very important too. I also would say number four, are they aware of what's happening right now in the news because I know we can use words like fake news. know that AI has allowed us to have ways to produce different types of news articles, but your reputation is everything. For almost every assessment I've ever done, I always stress to people that reputational damage has downstream effects. The truth or falsity of it we won't even get into. You don't want to be part of something inadvertently and what are you going to say and your positioning for not having governance if and when something were to happen that you became part of, whether it was an API connection, a breach, data dissemination, these things can and will happen. I think that'd be important for the right subject matter expertise to get involved. I'm certainly not suggesting that the individuals that you're working with don't have subject matter expertise in other areas, but it sounds like they do need people with more transferable or adequate knowledge of the governance space. Dr. David: Great. yeah, I I have a few things that overlap with that. So I think first is as you said, visibility. Where is AI in the organization? What are the specific use cases? Is it possible that we have personal data or proprietary business information that is going into these systems? Do we even have a a set of business licenses so that we could have some type of visibility into data flows there? you know. So do we have a sense of shadow AI and whether or not our employees are putting things into personal commercial subscriptions, which you know are are going off to to to large companies, own infrastructure, right? Are are we losing data that could potentially get incorporated into future training? and and those are just examples of, you know, a handful of risks. I would, you know, encourage people to consider the the spectrum of risks that are are available. So if we think of, you know, our our pestle assessment or pestle er, I don't know if you were on the call for that. So political, economic, social, technological, legal, environmental, and reputation. And just walk down each one of those. how do Emikly Dorlack: yes. Dr. David: how does how does not having a governance, what negative risks does that involve? I think it's also very important, and this is When it comes to the vision of the podcast and the collective, I think it's important as AI governance professionals that we are not seen as the department of no. We don't want to position ourselves Emikly Dorlack: Right? Dr. David: or make it so that in every meeting, we don't want to always be saying, you can't do this or you have to slow down. Because that type of framing. It it frames us as constraints on the business. And it it seems like if this is the response from an individual, an organization, or a number of organizations, that's their default mindset is that, you know, GRC or AI governance, this is a constraint. So in addition to making sure organizations are educated on negative risk, we also need to understand what positive risk is. And this is where the you know the ISO risk management frameworks, like 31,000. Are very useful because they're not focusing just on negative risks, but positive as well. and I believe the the NIST AI RMF has a point on this as well. So, Emikly Dorlack: It does. Dr. David: you know, you know, competitive disadvantages for not deploying it. what are our missed opportunities? Are we missing building out efficiencies? And is there something that we could do to Just to make employees' lives better, right? There's this whole other basket of positive risks that we could potentially suffer if we do not deploy the technology. And AI governance is doing that as well. a couple other things like also AI literacy going to positive risks. How do we make sure that employees are leveraging the tools to the best of their ability to again do more with less? tracking return on investment. And I I already said a little bit there on compliance with laws and Regulations. One thing that one of our early guests is going to help us understand is the patchwork of AI state-level laws. Without federal regulation, there is already a lot of existing law, law that has been around for decades, as well as a lot of new laws that have been signed over the last year or two that organizations are getting in trouble for. And so there's a lot out there that could. Jeopardize an organization, putting them in risk when they fail to consider AI in their organization. Emikly Dorlack: agree I would agree with that too and I always say, I remember sitting in one of my advanced torts classes, and now I'm getting nostalgic, and thinking, I love product liability and I love the consumer aspect of technology right now. And developing a mindset of AI being a product and treating it like that hits home exactly what you're saying, Doctor David, is people that sometimes run this assumption of, we have some state laws or there's kind of a federal period of Freedom, if you will, and less regulation, more hands-off, that does not absolve you of other existing requirements your organization may have. Dr. David: Yeah. th so the FTC has recently launched investigations sending civil investigative demands to anthropic open AI, Emikly Dorlack: Right. Dr. David: all under section five of the FTC Act, right? Unfair or deceptive acts and practices. So Emikly Dorlack: Understandably. Yep. Dr. David: that was law, what, in nineteen fourteen? So that's well over a hundred years old. And yeah, Emikly Dorlack: I think it's over a hundred. Dr. David: yeah. So j just an example there. All right, we've got a question here didn't get the result they wanted on their first attempt at the AI G P scoring low in domain four. ACET the master class and official practice exam quizzes, best way to refocus. So so typically when when I'm speaking with folks that fall short on their first attempt, I like to look at the the full score report. So please email that to me. if if you're comfortable sharing it, dr david at drdavidprivacy.com and I can get back to you with more tailored guidance. That's important just to see what the distribution is. The three mistakes, and I've written and talked a lot about this, so so I'll do this briefly. The top three mistakes that people make on AIGP, the number one mistake is that they don't pay enough attention to domain one. Domain one, even though it's not the most heavily weighted, is the most important domain. It's the technological foundations that undergird the entire exam. You may have a question that's testing you on domain two, three, or four, but there's there's very likely to be some. Technical component to that question that you have to unlock in order to answer a legal question or an AI development lifecycle question. And if you can't unlock that technical component, then you know chances are good you may not identify the right question to answer or the best answer that's available. so for example, in domain one, if you look at the 101 AIGP key terms glossary, 70 of those terms come from domain one. And then you have the other three domains for 31 terms. So, two-thirds of that terminology is all in domain one, which just it it it highlights its importance. the second mistake that people often make is that they they they do not commit to memory the different processes and and workflows, you you're very likely on this exam to get questions that ask you about the next best step or, you know, where you are in the AI development lifecycle. And if you haven't committed that sequencing to memory, that's gonna jam you up as well. And then the very third, the third most common mistake people make is they're just not prepared strategically for the exam. For how the questions are written, the different types of question formats that you're likely to get. This is just as much an English and reading exam as it is a subject matter exam. So you absolutely have to be prepared for this strategically. I want to provide two free resources that are both available on my website. So the first one is an IAPP exam secrets class. Even if you've done an IAPP exam before. and especially if this is your first exam, you should do that. It's only 53 minutes, it's absolutely free. If you go to drdavidprivacy.com, just you can register for an account and you can take that course. The the second resource that's available, and this is relatively new, I just launched it in the last week or so, is a free AIGP key terms quiz. It tests you on the 101 key terms for the exam, so you can see where you land with those. You need to be 100%. All right. You need to get 100% on that exam, or let's say over 90%. so it's four exams. Each one has 25 to 26 questions, and they're grouped thematically. but go over those, read over the answer explanations. That's that that's the the best advice that that I have for that. and once I see your scores, I can probably give you, you know, a little more feedback on that. Alrighty, question, Emikly Dorlack: Right age. Dr. David: let me see. Faisal is asking, how do I access those resources that are being discussed? so FISEL, if you could tell us a little more about the the resources that you're referring to, we can point you in the right direction. And then Chaya has another question here. ask these questions and what this company ended up doing, creating a two page AI literacy using Claude and called it AI literacy. It yeah, well, Emikly Dorlack: Mm. Dr. David: y I I I mean like it it it's at some at at some point, Emikly Dorlack: Sorry. Dr. David: you know, the the people running organizations need to be adults. and i if someone doesn't want to listen, I'm afraid there's probably not a whole lot you can do other than what we've said to to to get them to listen. But Emily, I don't know. Do you do you have anything more there? You're you're trying to sell someone on AI governance and and it sounds like they just maybe they had a bad experience in the past. I I don't know. Emikly Dorlack: be careful about how I frame this. I think as an individual, maybe just take in everything you know about the organization's culture. You know, I think sometimes governance also is unpacking areas of an organization that might not at the surface show red flags and whether that's your data quality, whether it's broken systems. I think I can say this from experience. Organizations, whether intentionally or inadvertently, they do have broken system broken systems, processes. They may even have data issues. and also that's complicated by technology, that's compounded by all these organizational shifts. We've seen so many reductions in force. And if you're gonna approach AI governance, you're gonna unpack things you might not want to see or or deal with. And so maybe, you know, in your own personal time, unpacking or trying to identify what those might be. Because we've talked about how you can have a conversation with leaders, but then just trying to really be honest with yourself about the culture of your organization will help. We are we're gonna repeatedly talk about building a culture of trust. How do you build that culture given the current parameters you're given, right? And you're identifying one. How do I get leadership buy-in? Dr. David: Yeah. You know, another thing that I'm thinking, and and this requires the organization's cooperation, but typically engagements will start with a gap assessment and you know you will, you know, have a framework like like NIST Emikly Dorlack: Sure. Dr. David: that you're running that gap assessment across. And you know, if you're if you're starting out and you're just trying to identify policies and and other important documents and the organizations don't have those, you know, then you can frame that, you know, if there were some type of regulatory Action, you know, if regulators or auditors show up and they're Emikly Dorlack: Exactly. Dr. David: they're asking for this and it's not written down, then I mean you fail right away before you even get started. So I I think trying to to frame, you know, what concrete exam or consequences would look like due to not having compliance or processes in place. one thing that came up in one of our discussions earlier, also on this topic of buy-in, was going to something like the The AI incident database. Or the MIT risk repository, and Emikly Dorlack: That's right. Dr. David: you know, finding, you know, let's say that you you know what industry this company is that you're speaking in or that you're speaking to, identifying incidents in their space and how that has created, you know, some some material harm to the business, to their reputation, you know, regulatory harm, or or so forth. And, you know, framing it in a sense that, This is what's happened to other people in your space for this type of issue, right? For some type of harmful output or you know, misuse of the system or you know unauthorized disclosure of personal data and and so forth. finding concrete examples that have have been in the media, you know, that that may be helpful to to continue the conversation. Emikly Dorlack: I'd I'd one more thought on this as as we're talking and maybe that's my new tagline. I have one more thought. in all seriousness though, I came from a a different end of the risk scope, right? And so what I love about these conversations, Kyle, too, is seeing what it's like in environments that don't have that scrutiny and I I think having things on lockdown, right? Outright prohibiting any person internally from downloading Deep Seek or accessing X, Y, or Z list. I think that's important to consider too, but when we talk about culture, how is it going to be perceived as your organization, especially when there's this huge pitch for AI having this ability to make us better at our work and that it's emerging in different spaces and everyone's using it? Because then you have that economic argument too that if all of the individuals, clients, vendors that you're working with are using it. that you should be using it too or aware of the implications from a governance perspective where wherever you flow industry-wise. Dr. David: Yeah. So someone was asking for, you know, where our NIST project and working group is. It's at Emikly Dorlack: Mm-hmm. Dr. David: it's within the AI governance collective. So go to AIGov Collective.com, Emikly Dorlack: Yeah. Dr. David: All right, let me see. We probably got let's I think we got one more question. So Emikly Dorlack: Sure. Dr. David: it looks like Caro from Germany, where the EU AI Act is central to AI governance. How deeply does the exam prep cover EU AI Act, particularly in practical implications for enterprise adoption? Yeah. So Emikly Dorlack: question. Dr. David: there's there's a lot that I would love to add to the AI GP certification masterclass. there there are a lot of important issues that deserve a lot more attention. The the challenge is the objective of the course is to prepare you for the exam. And I'm I'm very conscious about providing you exactly what you need to prepare for and and pass the exam on your first attempt. that does include like building up to certain topics. I Try and frame all of my courses so that they're suitable for beginners. That means that I'm putting in some introductory material to to build you up so that you can understand a particular concept. the EU AI Act does loom large in my course and on the exam, but it it I I do pull in practical applications where it helps to demonstrate or or make a per a particular point concrete. But would you be ready to go in and and implement or do a gap assessment for an organization based just on your preparation for the exam? you would you would need some help, right? You would you would definitely want to make sure that you know where to go to find the legislative text and and how to read through it. so this is it's not an EU AI Act implementation course. It's an AI G Pur. certification exam course. That that's the objective. So there is a good amount of content on the EU AI Act, but the objective is to prepare you for the exam. and and that does provide a foundation for actually doing the work, but you would you would definitely need more training. okay. Yes, thank you, Caro. I appreciate it. So they put the the link in there for us, the right one. All Emikly Dorlack: Thank you. Great. Dr. David: right. so thank you everyone so much for for hanging out for our very first episode live here. I think Emily, I th I think it was okay. I I think we did a good job. You dropped. your your mic dropped, which is a very fitting way to we there you go, you're back. We we will we will work that out for the next time. So th this was a good dry run. Emikly Dorlack: back. We will indeed. Dr. David: Yeah. All right. thanks so much, everyone. really enjoyed this and we we will have a lot more to share. where you can subscribe to the podcast, you know, how often to to expect it. We plan on doing this weekly the the goal is to have our first full episode along with our first interview guest on November tenth. thanks so much, everyone. Have a fantastic Halloween and we'll see you soon. Bye. Emikly Dorlack: Bye everyone.