WEBVTT - Smart Talks with IBM: Responsible AI: Why Businesses Need Reliable AI Governance 0:00:00.120 --> 0:00:02.840 Hey everyone, it's Robert and Joe here. Today we've got 0:00:02.840 --> 0:00:04.720 something a little bit different to share with you. It 0:00:04.840 --> 0:00:08.000 is a new season of the Smart Talks with IBM 0:00:08.119 --> 0:00:09.119 podcast series. 0:00:09.600 --> 0:00:11.680 Today we are witnessed to one of those rare moments 0:00:11.680 --> 0:00:14.360 in history, the rise of an innovative technology with the 0:00:14.360 --> 0:00:18.680 potential to radically transform business and society forever. The technology, 0:00:18.760 --> 0:00:22.200 of course, is artificial intelligence, and it's the central focus 0:00:22.239 --> 0:00:24.800 for this new season of Smart Talks with IBM. 0:00:25.320 --> 0:00:28.400 Join hosts from your favorite Pushkin podcasts as they talk 0:00:28.480 --> 0:00:31.640 with industry experts and leaders to explore how businesses can 0:00:31.680 --> 0:00:35.360 integrate AI into their workflows and help drive real change 0:00:35.400 --> 0:00:38.160 in this new era of AI. And of course, host 0:00:38.280 --> 0:00:40.440 Malcolm Gladwell will be there to guide you through the 0:00:40.479 --> 0:00:42.640 season and throw in his two cents as well. 0:00:43.120 --> 0:00:46.000 Look out for new episodes of Smart Talks with IBM 0:00:46.400 --> 0:00:49.519 every other week on the iHeartRadio app, Apple Podcasts, or 0:00:49.560 --> 0:00:53.360 wherever you get your podcasts, and learn more at IBM 0:00:53.479 --> 0:00:55.480 dot com slash smart Talks. 0:00:58.000 --> 0:01:02.040 Hello, Hello, Welcome to Smart with IBM, a podcast from 0:01:02.080 --> 0:01:08.440 Pushkin Industri's iHeartRadio and IBM. I'm Malcolm Glabwell. This season 0:01:08.480 --> 0:01:12.840 we're continuing our conversation with new creators visionaries who are 0:01:12.880 --> 0:01:16.399 creatively applying technology in business to drive change, but with 0:01:16.480 --> 0:01:21.399 a focus on the transformative power of artificial intelligence and 0:01:21.480 --> 0:01:24.600 what it means to leverage AI as a game changing 0:01:24.720 --> 0:01:29.720 multiplier for your business. Our guest today is Christina Montgomery, 0:01:30.120 --> 0:01:34.200 IBM's Chief Privacy and Trust Officer. She's also chair of 0:01:34.240 --> 0:01:39.679 IBM's AI Ethics Board. In addition to overseeing IBM's privacy policy, 0:01:40.040 --> 0:01:45.039 a core part of Christina's job involves AI governance, making 0:01:45.080 --> 0:01:48.880 sure the way AI is used complies with the international 0:01:49.200 --> 0:01:55.040 legal regulations customized for each industry. In today's episode, Christina 0:01:55.080 --> 0:01:59.920 will explain why businesses need foundational principles when it comes 0:01:59.960 --> 0:02:04.800 to using technology, why AI regulation should focus on specific 0:02:05.000 --> 0:02:09.280 use cases over the technology itself, and share a little 0:02:09.280 --> 0:02:14.920 bit about her landmark congressional testimony. Last May, Christina spoke 0:02:14.960 --> 0:02:18.640 with doctor Lori Santos, host of the Pushkin podcast The 0:02:18.680 --> 0:02:24.160 Happiness Lab, a cognitive scientist and psychology professor at Yale University. 0:02:24.480 --> 0:02:29.840 Laurie is an expert on human happiness and cognition. Okay, 0:02:30.360 --> 0:02:31.639 let's get to the interview. 0:02:34.200 --> 0:02:36.280 So Christina, I'm so excited to talk to you today. 0:02:36.440 --> 0:02:38.640 So let's start by talking a little bit about your 0:02:38.720 --> 0:02:41.400 role at IBM. What does a Chief Privacy and Trust 0:02:41.440 --> 0:02:42.480 officer actually do. 0:02:43.000 --> 0:02:46.760 It's a really dynamic profession and it's not a new profession, 0:02:47.320 --> 0:02:50.200 but the role has really changed. I mean, my role 0:02:50.320 --> 0:02:54.040 today is broader than just helping to ensure compliance with 0:02:54.160 --> 0:02:58.800 data protection laws globally. I'm also responsible for AI governance. 0:02:58.840 --> 0:03:01.799 I co chair or AI Ethics Board here at IBM, 0:03:02.120 --> 0:03:05.160 and for data clearance and data governance as well for 0:03:05.240 --> 0:03:08.800 the company. So I have both a compliance aspect to 0:03:08.840 --> 0:03:11.560 my role, really important on a global basis, but also 0:03:12.080 --> 0:03:17.000 help the business to competitively differentiate because really trust is 0:03:17.000 --> 0:03:20.720 a strategic advantage for IBM and a competitive differentiator as 0:03:20.720 --> 0:03:24.640 a company that's been responsibly managing the most sensitive data 0:03:24.680 --> 0:03:27.320 for our clients for more than a century now and 0:03:27.440 --> 0:03:30.160 helping to usher new technologies into the world with trust 0:03:30.160 --> 0:03:33.560 and transparency. And so that's also a key aspect of 0:03:33.600 --> 0:03:34.040 my role. 0:03:34.800 --> 0:03:37.000 And so you joined us here on smart Talks back 0:03:37.000 --> 0:03:39.240 in twenty twenty one and you chatted with us about 0:03:39.240 --> 0:03:42.920 IBM's approach of building trust and transparency with AI. And 0:03:43.040 --> 0:03:45.080 that was only two years ago, but it almost feels 0:03:45.120 --> 0:03:47.440 like in eternity has happened in the field of AI 0:03:47.600 --> 0:03:50.600 since then, and so I'm curious how much has changed 0:03:50.640 --> 0:03:53.120 since you were here last time. We're the things you 0:03:53.160 --> 0:03:55.840 told us before, you know, are they still true? How 0:03:55.880 --> 0:03:56.680 are things changing? 0:03:57.080 --> 0:04:00.560 You're absolutely right, it feels like the world has changed 0:04:00.640 --> 0:04:04.320 really in the last two years. But the same fundamental 0:04:04.360 --> 0:04:09.280 principles and the same overall governance apply to IBM's program 0:04:09.840 --> 0:04:14.200 for Data Protection and Responsible AI that we talked about 0:04:14.240 --> 0:04:16.920 two years ago, and not much has changed there from 0:04:16.960 --> 0:04:20.120 our perspective. And the good thing is we've put these 0:04:20.160 --> 0:04:24.599 practices and this governance approach into place, and we have 0:04:24.839 --> 0:04:28.480 an established way of looking at these emerging technologies. As 0:04:28.520 --> 0:04:31.920 the technology evolves, the tech is more powerful, for sure, 0:04:32.080 --> 0:04:36.120 foundation models are vastly larger and more capable, and our 0:04:36.200 --> 0:04:39.640 creating in some respects new issues, but that just makes 0:04:39.640 --> 0:04:41.440 it all the more urgent to do what we've been 0:04:41.480 --> 0:04:45.080 doing and to put trust and transparency into place across 0:04:45.120 --> 0:04:47.760 the business to be accountable to those principles. 0:04:48.960 --> 0:04:51.120 And so our conversation today is really centered around this 0:04:51.279 --> 0:04:54.520 need for new AI regulation, and part of that regulation 0:04:54.640 --> 0:04:57.520 involves the mitigation of bias and This is something I 0:04:57.520 --> 0:05:00.200 think about a ton as a psychologist, right, you know, 0:05:00.600 --> 0:05:03.719 like my students and everyone who's interacting with AI is 0:05:03.760 --> 0:05:06.520 assuming that the kind of knowledge that they're getting from 0:05:06.640 --> 0:05:09.359 this kind of learning is accurate, right, But of course 0:05:09.400 --> 0:05:11.680 AI is only as good as the knowledge that's going in. 0:05:11.720 --> 0:05:14.080 And so talk to me a little bit about like 0:05:14.560 --> 0:05:17.039 why bias occurs in AI and the level of the 0:05:17.080 --> 0:05:18.520 problem that we're really dealing with. 0:05:19.360 --> 0:05:22.400 Yeah, Well, obviously AI is based on data, right, It's 0:05:22.760 --> 0:05:27.760 trained with data, and that data could be biased in 0:05:27.760 --> 0:05:30.200 and of itself, and that's where issues could come up. 0:05:30.279 --> 0:05:32.160 They come up in the data, they could also come 0:05:32.240 --> 0:05:35.760 up in the output of the models themselves. So it's 0:05:35.880 --> 0:05:40.640 really important that you build bias consideration and bias testing 0:05:40.760 --> 0:05:44.160 into your product development cycle. And so what we've been 0:05:44.200 --> 0:05:46.760 thinking about here at IBM and doing we had some 0:05:46.839 --> 0:05:49.799 of our research teams delivered some of the very first 0:05:49.800 --> 0:05:53.159 toolkits to help detect bias years ago now, right, and 0:05:53.200 --> 0:05:56.840 deployed them to open source. And we have put into 0:05:56.880 --> 0:06:00.200 place for our developers here at IBM and Ethics by 0:06:00.240 --> 0:06:03.480 Design playbook that's sort of a step by step approach 0:06:04.200 --> 0:06:10.720 which also addresses very fully bias considerations, and we provide 0:06:10.760 --> 0:06:13.960 not only like here's a point when you should test 0:06:14.000 --> 0:06:16.400 for it and you consider it in the data, you 0:06:16.440 --> 0:06:18.159 have to measure it both at the data and the 0:06:18.200 --> 0:06:21.720 model level or the outcome level, and we provide guidance 0:06:21.720 --> 0:06:24.400 with respect to what tools can best be used to 0:06:24.440 --> 0:06:27.920 accomplish that. So it's a really important issue. It's one 0:06:27.960 --> 0:06:30.920 you can't just talk about. You have to provide essentially 0:06:31.000 --> 0:06:34.279 the technology and the capabilities and the guidance to enable 0:06:34.320 --> 0:06:35.480 people to test for it. 0:06:36.200 --> 0:06:38.880 Recently, you had this wonderful opportunity to head to Congress 0:06:38.920 --> 0:06:41.960 to talk about AI, and in your testimony before Congress, 0:06:42.000 --> 0:06:45.000 you mentioned that it's often said that innovation moves too 0:06:45.040 --> 0:06:47.640 fast for government to keep up. And this is something 0:06:47.640 --> 0:06:49.800 that I also worry about as a psychologist. Right our 0:06:49.839 --> 0:06:52.760 policy makers really understanding the issues that they're dealing with, 0:06:53.200 --> 0:06:55.440 And so I'm curious how you're approaching this challenge of 0:06:55.480 --> 0:06:57.920 adapting AI policies to keep up with the sort of 0:06:58.000 --> 0:07:00.760 rapid pace of all the advancements were in the AI 0:07:00.839 --> 0:07:01.880 technology itself. 0:07:02.839 --> 0:07:06.039 I think it's really critically important that you have foundational 0:07:06.120 --> 0:07:11.560 principles that applied to not only how you use technology, 0:07:11.600 --> 0:07:13.080 but whether you're going to use it in the first 0:07:13.120 --> 0:07:14.760 place and where you're going to use and apply it 0:07:14.760 --> 0:07:18.840 across your company. And then your program from a governance perspective, 0:07:18.920 --> 0:07:21.440 has to be agile. It has to be able to 0:07:21.520 --> 0:07:27.240 address emerging capabilities, new training methods, etc. And part of 0:07:27.280 --> 0:07:31.800 that involves helping to educate and instill and empower a 0:07:31.840 --> 0:07:35.640 trustworthy culture at a company so you can spot those 0:07:35.680 --> 0:07:37.920 issues so you can ask the right questions at the 0:07:38.000 --> 0:07:41.200 right time if you try. We talked about during the 0:07:41.240 --> 0:07:45.440 Senate hearing, and IBM's been talking for years about regulating 0:07:45.760 --> 0:07:48.920 the use, not the technology itself, because if you try 0:07:48.920 --> 0:07:52.760 to regulate technology, you're very quickly going to find out 0:07:53.160 --> 0:07:56.120 regulation will absolutely never keep up with that. 0:07:56.600 --> 0:07:58.800 And so in your testimony to Congress, you also talked 0:07:58.840 --> 0:08:02.040 about this idea of a precision regulation approach for AI. 0:08:02.680 --> 0:08:04.880 Tell me more about this. What is a precision regulation 0:08:04.960 --> 0:08:06.960 approach and why could that be so important. 0:08:07.240 --> 0:08:10.200 It's funny because I was able to share with Congress 0:08:10.840 --> 0:08:14.640 our precision regulation point of view in twenty twenty three, 0:08:15.000 --> 0:08:17.840 but that precision regulation point of view was published by 0:08:17.920 --> 0:08:22.600 IBM in twenty twenty so we have not changed our 0:08:22.720 --> 0:08:27.640 position that you should apply the tightest controls, the strictest 0:08:27.680 --> 0:08:31.920 regulatory requirements to the technology where the end use and 0:08:32.040 --> 0:08:35.960 risk of societal harm is the greatest. So that's essentially 0:08:36.000 --> 0:08:38.960 what it is. There's lots of AI technology that's used 0:08:39.000 --> 0:08:42.640 today that doesn't touch people, that's very low risk in nature. 0:08:43.160 --> 0:08:46.600 And even when you think about AI that delivers a 0:08:46.720 --> 0:08:52.160 movie recommendation versus AI that is used to diagnose cancer, right, 0:08:52.200 --> 0:08:55.680 there's very different implications associated with those two uses of 0:08:55.720 --> 0:09:00.480 the technology. And so essentially what precision regulation it is 0:09:00.480 --> 0:09:03.400 is apply different rules to different risks, right, more stringent 0:09:03.480 --> 0:09:07.120 regulation to the use cases with the greatest risk. And 0:09:07.160 --> 0:09:10.800 then also we build that out calling for things like 0:09:10.840 --> 0:09:15.440 transparency you see it today with content right, misinformation and 0:09:15.480 --> 0:09:19.600 the like. We believe that consumers should always know when 0:09:19.600 --> 0:09:22.640 they're interacting with an AI system, So be transparent, don't 0:09:22.720 --> 0:09:27.480 hydror AI clearly define the risks. So as a country, 0:09:27.760 --> 0:09:30.640 we need to have some clear guidance right and globally 0:09:30.679 --> 0:09:34.360 as well in terms of which uses of AI or 0:09:34.480 --> 0:09:38.880 higher risk where will apply higher and stricter regulation and 0:09:39.080 --> 0:09:41.600 have sort of a common understanding of what those high 0:09:41.679 --> 0:09:45.440 risk uses are and then demonstrate the impact in the 0:09:45.480 --> 0:09:50.079 cases of those higher risk uses. So companies who are 0:09:50.160 --> 0:09:54.600 using AI in spaces where they can impact people's legal rights, 0:09:54.640 --> 0:09:59.000 for example, should have to conduct an impact assessment that 0:09:59.080 --> 0:10:02.880 demonstrates that the technology isn't biased. So we've been pretty 0:10:02.880 --> 0:10:06.559 clear about apply that the most stringent regulation to the 0:10:06.640 --> 0:10:08.000 highest risk uses of AI. 0:10:09.760 --> 0:10:12.040 And so so far we've been talking about your congressional 0:10:12.040 --> 0:10:14.880 testimony in terms of, you know, the specific content that 0:10:14.920 --> 0:10:17.400 you talked about, But I'm just curious on a personal level, 0:10:17.720 --> 0:10:19.839 you know, what was that like right like, right now, 0:10:19.840 --> 0:10:22.120 it feels like at a policy level, like there's a 0:10:22.200 --> 0:10:24.559 kind of fever pitch going on with AI right now. 0:10:24.800 --> 0:10:26.240 You know, what did that feel like to kind of 0:10:26.320 --> 0:10:28.760 really have the opportunity to talk to policy makers and 0:10:28.800 --> 0:10:31.800 sort of influence what they're thinking about AI technologies like 0:10:31.880 --> 0:10:32.840 in the coming century. 0:10:32.880 --> 0:10:35.800 Perhaps I was really an honor to be able to 0:10:35.880 --> 0:10:38.840 do that and to be one of the first set 0:10:38.880 --> 0:10:42.960 of invitees to the first hearing. And what I learned 0:10:43.000 --> 0:10:46.160 from it essentially is, you know, really two things. The 0:10:46.200 --> 0:10:50.079 first is really the value of authenticity. So both as 0:10:50.240 --> 0:10:54.200 an individual and as a company, I was able to 0:10:54.240 --> 0:10:56.800 talk about what I do, you know, I need a 0:10:56.800 --> 0:11:00.640 lot of advanced prep, right, I talked about my job 0:11:00.720 --> 0:11:04.520 is what IBM has been putting in place for years now. 0:11:05.080 --> 0:11:08.960 So this isn't about creating something. This was just about 0:11:09.000 --> 0:11:11.320 showing up and being authentic. And we were invited for 0:11:11.360 --> 0:11:14.200 a reason. We were invited because we were one of 0:11:14.200 --> 0:11:18.960 the earliest companies in the AI technology space. We're the 0:11:19.000 --> 0:11:24.439 oldest technology company and we are trusted and that's an honor. 0:11:24.600 --> 0:11:26.840 And then the second thing I came away with was 0:11:26.880 --> 0:11:29.240 really how important this issue is to society. I don't 0:11:29.240 --> 0:11:34.720 think I appreciated it as much until following that experience. 0:11:35.360 --> 0:11:39.160 I had outreached from colleagues I hadn't worked with for years. 0:11:39.240 --> 0:11:41.440 I had outreach from family members who heard me on 0:11:41.480 --> 0:11:44.600 the radio. You know, my mother and my mother in 0:11:44.679 --> 0:11:47.760 law and my nieces and nephews and my friends of 0:11:47.800 --> 0:11:49.640 my kids were all like, oh, I get it, I 0:11:49.679 --> 0:11:50.360 get what you do. 0:11:50.440 --> 0:11:50.640 Now. 0:11:50.679 --> 0:11:53.160 Wow, that's pretty cool, you know. So that was really 0:11:53.920 --> 0:11:57.080 probably the best and most impactful takeaway that I had. 0:11:57.559 --> 0:12:01.760 The mass adoption of generative AI at breakneck speed has 0:12:01.800 --> 0:12:06.160 spurred societies and governments around the world to get serious 0:12:06.640 --> 0:12:11.960 about regulating AI. For businesses, compliance is complex enough already, 0:12:12.120 --> 0:12:15.480 but throw an ever involving technology like AI into the mix, 0:12:15.880 --> 0:12:22.160 and compliance itself becomes an exercise in adaptability. As regulators 0:12:22.160 --> 0:12:26.720 seek greater accountability in how AI is used, businesses need 0:12:26.760 --> 0:12:31.360 help creating governance processes comprehensive enough to comply with the 0:12:31.440 --> 0:12:34.839 law but agile enough to keep up with the rapid 0:12:34.920 --> 0:12:40.040 rate of change in AI development. Regulatory scrutiny isn't the 0:12:40.120 --> 0:12:45.240 only consideration either responsible AI governance of business's ability to 0:12:45.400 --> 0:12:50.199 prove its AI models are transparent and explainable is also 0:12:50.320 --> 0:12:55.480 key to building trust with customers, regardless of industry. In 0:12:55.520 --> 0:12:59.240 the next part of their conversation, Laurie asked Christina what 0:12:59.320 --> 0:13:04.320 businesses SHO should consider when approaching AI governance. Let's listen. 0:13:05.160 --> 0:13:07.559 So it's a particular role that businesses are playing in 0:13:07.679 --> 0:13:10.479 AI governance, Like why is it so critical for businesses 0:13:10.480 --> 0:13:11.480 to be part of this? 0:13:12.040 --> 0:13:16.480 So I think it's really critically important that businesses understand 0:13:17.040 --> 0:13:19.959 the impacts that technology can have, both in making them 0:13:19.960 --> 0:13:23.800 better businesses, but the impacts that those technologies can have 0:13:24.280 --> 0:13:29.400 on the consumers that they are supporting. You know, businesses 0:13:29.520 --> 0:13:34.120 need to be deploying AI technology that is in alignment 0:13:34.280 --> 0:13:36.160 with the goals that they set for it, and that 0:13:36.280 --> 0:13:39.160 can be trusted. I think for us and for our clients, 0:13:39.480 --> 0:13:42.400 a lot of this comes back to trust in tech. 0:13:42.840 --> 0:13:48.520 If you deploy something that doesn't work, that hallucinates, that discriminates, 0:13:49.000 --> 0:13:53.120 that isn't transparent, where decisions can't be explained, then you 0:13:53.280 --> 0:13:56.920 are going to very rapidly erode the trust at best 0:13:57.080 --> 0:14:00.280 right of your clients and at worst for yourself if 0:14:00.280 --> 0:14:02.720 you're going to create legal and regulatory issues for yourself 0:14:02.760 --> 0:14:06.680 as well. So trusted technology is really important, and I 0:14:06.679 --> 0:14:08.880 think there's a lot of pressure on businesses today to 0:14:08.920 --> 0:14:11.600 move very rapidly and adopt technology. But if you do 0:14:11.640 --> 0:14:14.800 it without having a program of governance in place, you're 0:14:14.840 --> 0:14:16.800 really risking eroding that trust. 0:14:16.920 --> 0:14:19.120 So this is really where I think a strong AI 0:14:19.200 --> 0:14:22.760 governance comes in. Talk about from your perspective, how this 0:14:22.840 --> 0:14:26.720 really contributes to maintaining the trust that customers and stakeholders 0:14:26.760 --> 0:14:27.920 have in these technologies. 0:14:28.640 --> 0:14:31.200 Yeah. Absolutely, I mean you need to have a governance 0:14:31.240 --> 0:14:35.240 program because you need to understand that the technology, particularly 0:14:35.240 --> 0:14:39.840 in the AI space, that you are deploying, is explainable. 0:14:39.880 --> 0:14:44.240 You need to understand why it's making decisions and recommendations 0:14:44.280 --> 0:14:45.760 that it's making, and you need to be able to 0:14:45.760 --> 0:14:48.160 explain that to your consumers. I mean, you can't do 0:14:48.240 --> 0:14:50.320 that if you don't know where your data is coming from, 0:14:50.320 --> 0:14:52.800 what data are you using to train those models, if 0:14:52.800 --> 0:14:56.600 you don't have a program that manages the alignment of 0:14:56.640 --> 0:14:59.840 your AI models over time to make sure as a 0:15:00.160 --> 0:15:04.440 I learns and evolves over uses, which is in large 0:15:04.480 --> 0:15:08.760 part what makes it so beneficial that it stays in 0:15:08.800 --> 0:15:12.200 alignment with the objectives that you set for the technology 0:15:12.200 --> 0:15:15.840 over time. So you can't do that without a robust 0:15:16.160 --> 0:15:20.000 governance process in place. So we work with clients to 0:15:20.120 --> 0:15:22.920 share our own story here at IBM in terms of 0:15:22.960 --> 0:15:25.720 how we put that in place, but also in our 0:15:25.720 --> 0:15:31.360 consulting practice to help clients work with these new generative 0:15:31.400 --> 0:15:34.560 capabilities and foundation models and the like in order to 0:15:34.760 --> 0:15:36.680 put them to work for their business in a way 0:15:36.720 --> 0:15:39.480 that's going to be impactful to that business but at 0:15:39.480 --> 0:15:41.000 the same time be trusted. 0:15:41.400 --> 0:15:43.280 So now I wanted to turn a little bit towards 0:15:43.280 --> 0:15:47.120 watsonex governance, and so IBM recently announced their AI platform, 0:15:47.160 --> 0:15:50.880 Watson X, which will include a governance component. Could you 0:15:50.920 --> 0:15:53.520 tell us a little more about watsonex dot Governance. 0:15:53.880 --> 0:15:56.120 Yeah, I mean before I do that, I'll just back 0:15:56.200 --> 0:15:59.760 up and talk about the full platform and then lean 0:15:59.800 --> 0:16:02.120 into to watson x because I think it's important to 0:16:02.200 --> 0:16:08.080 understand the delivery of a full suite of capabilities, to 0:16:08.120 --> 0:16:11.240 get data, to train models, and then to govern them 0:16:11.440 --> 0:16:16.720 over their life cycle. All of these things are really important. 0:16:17.280 --> 0:16:20.360 From the onset you need to make sure that you have, 0:16:21.000 --> 0:16:25.240 you know, for our watsonex dot AI for example, that's 0:16:25.280 --> 0:16:29.840 the studio to train new foundation models and generative AI 0:16:29.920 --> 0:16:35.560 and machine learning capabilities, and we are populating that studio 0:16:35.920 --> 0:16:41.200 with some IBM trained foundation models which we're curating and 0:16:41.280 --> 0:16:45.040 tailoring more specifically for enterprises. So that's really important. It 0:16:45.040 --> 0:16:47.440 comes back to the point I made earlier about business 0:16:47.440 --> 0:16:52.200 trust and the need, you know, to have enterprise ready 0:16:53.240 --> 0:16:57.360 technologies in the AI space. And then the watsonex dot 0:16:57.480 --> 0:17:01.040 data is a fit for purpose data or a data lake, 0:17:01.640 --> 0:17:06.120 and then watsonex dot gov. So that's a particular component 0:17:06.280 --> 0:17:10.240 of the platform that my team and the AI Ethics 0:17:10.280 --> 0:17:14.359 Board has really worked closely with the product team on developing, 0:17:14.400 --> 0:17:17.240 and we're using it internally here in the Chief Privacy 0:17:17.280 --> 0:17:21.640 Office as well to help us govern our own uses 0:17:21.720 --> 0:17:26.760 of AI technology and our compliance program here, and it 0:17:26.880 --> 0:17:32.240 essentially helps to notify you if a model becomes biased 0:17:32.280 --> 0:17:35.440 or gets out of alignment as you're using it over time. 0:17:36.000 --> 0:17:38.639 So companies are going to need these capabilities. I mean 0:17:38.680 --> 0:17:42.840 they need them today to deliver technologies with trust. They'll 0:17:42.880 --> 0:17:46.159 need them tomorrow to comply with regulation, which is on 0:17:46.240 --> 0:17:47.359 the horizon. 0:17:47.080 --> 0:17:49.960 And I think compliance becomes even more complex when you 0:17:50.000 --> 0:17:54.120 consider international data protection laws and regulations. Honestly, I don't 0:17:54.160 --> 0:17:56.720 know how anyone on any company's legal team is keeping 0:17:56.800 --> 0:17:59.040 up with us these days. But my question for you 0:17:59.160 --> 0:18:02.760 is really how can businesses develop a strategy to maintain 0:18:02.800 --> 0:18:05.600 compliance and to deal with it in this ever changing landscape. 0:18:05.680 --> 0:18:09.600 It's increasingly more challenging. In fact, I saw statistic just 0:18:09.680 --> 0:18:14.280 this morning that the regulatory obligations on companies have increased 0:18:14.280 --> 0:18:18.160 something like seven hundred times in the last twenty years. 0:18:17.960 --> 0:18:22.560 So it really is a huge focus area for companies. 0:18:22.720 --> 0:18:25.600 You have to have a process in place in order 0:18:25.640 --> 0:18:28.199 to do that, and it's not easy, particularly for a 0:18:28.240 --> 0:18:32.120 company like IBM that it has a presence in over 0:18:32.119 --> 0:18:35.159 one hundred and seventy countries around the world. There is 0:18:35.200 --> 0:18:40.080 more than one hundred and fifty comprehensive privacy regulations. There 0:18:40.119 --> 0:18:44.640 are regulations of non personal data, there are AI regulations emerging, 0:18:45.600 --> 0:18:49.679 So you really need an operational approach to it in 0:18:49.800 --> 0:18:51.840 order to stay compliant. But one of the things we 0:18:51.840 --> 0:18:54.040 do is we set a baseline. And a lot of 0:18:54.040 --> 0:18:57.480 companies do this as well. So we define a privacy baseline, 0:18:57.520 --> 0:19:02.320 we define an AI baseline, and we ensure then as 0:19:02.359 --> 0:19:04.760 a result of that that there are very few deviances 0:19:04.840 --> 0:19:07.920 because it incorporates in that baseline. So that's one of 0:19:07.960 --> 0:19:10.119 the ways we do it. Other companies, I think are 0:19:10.200 --> 0:19:15.399 similarly situated in terms of doing that. But again, it 0:19:15.560 --> 0:19:18.240 is a real challenge for global companies. It's one of 0:19:18.240 --> 0:19:22.200 the reasons why we advocate for as much alignment as 0:19:22.280 --> 0:19:27.640 possible on the international realm as well as nationally here 0:19:27.680 --> 0:19:31.440 in the US, as much alignment as possible to make 0:19:31.680 --> 0:19:36.640 compliance easier for easier, and not just because companies want 0:19:36.680 --> 0:19:39.800 an easy way to comply. But the harder it is, 0:19:40.080 --> 0:19:43.960 the less likely there will be compliance. And it's not 0:19:44.080 --> 0:19:50.159 the objective of anybody, governments, companies, consumers to have to 0:19:50.320 --> 0:19:53.840 set legal obligations that companies simply can't meet. 0:19:54.080 --> 0:19:56.080 So what advice would you give to other companies who 0:19:56.119 --> 0:19:59.159 are looking to rethink or strengthen their approach to AI governance. 0:19:59.240 --> 0:20:02.600 Think you need to start with, as we did, foundational principles, 0:20:03.240 --> 0:20:06.679 and you need to start making decisions about what technology 0:20:06.680 --> 0:20:09.040 you're going to deploy and what technology you're not, What 0:20:09.040 --> 0:20:10.160 are you going to use it for, and what aren't 0:20:10.160 --> 0:20:11.560 you going to use it for? And then when you 0:20:11.560 --> 0:20:15.960 do use it, align to those principles. That's really important. 0:20:16.000 --> 0:20:20.560 Formalize a program, have someone within the organization, whether it's 0:20:20.600 --> 0:20:24.879 the chief privacy officer, whether it's some other role, a 0:20:24.960 --> 0:20:29.399 chief AI ethics officer, but have an accountable individual and 0:20:29.440 --> 0:20:33.639 accountable organization. Do a maturity assessment, figure out where you 0:20:33.720 --> 0:20:36.479 are and where you need to be, and really start, 0:20:36.680 --> 0:20:40.760 you know, putting it into place today. Don't wait for 0:20:41.200 --> 0:20:44.280 regulation to apply directly to your business because it'll be 0:20:44.359 --> 0:20:44.760 too late. 0:20:46.000 --> 0:20:49.400 So Smart Talks features new creators, these visionaries like yourself 0:20:49.400 --> 0:20:52.920 who are creatively applying technology in business to drive change. 0:20:53.160 --> 0:20:55.800 I'm curious if you see yourself as creative. 0:20:56.040 --> 0:20:59.320 You know, I definitely do. I mean, you need to 0:20:59.440 --> 0:21:04.080 be creative when you're working in an industry that evolves 0:21:04.119 --> 0:21:08.600 so very quickly. So you know, I started with IBM 0:21:08.880 --> 0:21:12.000 when we were primarily a hardware company, right, and we've 0:21:12.080 --> 0:21:15.480 changed our business so significantly over the years. And the 0:21:15.600 --> 0:21:20.040 issues that are raised with respect to each new technology, 0:21:20.080 --> 0:21:23.720 whether it be cloud, whether it be AI now where 0:21:23.720 --> 0:21:25.320 we're seeing a ton of issues, or you look at 0:21:25.320 --> 0:21:29.480 emergent issues in the space of things like neurotechnologies and 0:21:29.560 --> 0:21:36.160 quantum computers. You have to be strategic and you have 0:21:36.280 --> 0:21:39.520 to be creative and thinking about how you can adapt 0:21:40.160 --> 0:21:45.399 agilely quickly a company to an environment that is changing 0:21:45.440 --> 0:21:46.320 so quickly. 0:21:47.280 --> 0:21:50.240 And with this transformation happening at such a rapid pace. 0:21:50.440 --> 0:21:52.399 Do you think creativity plays a role in how you 0:21:52.440 --> 0:21:55.919 think about and implement specifically a trustworthy AI strategy. 0:21:58.160 --> 0:22:02.560 Yeah, think it does, because again, it comes back to 0:22:02.640 --> 0:22:05.879 these capabilities, and there are ways, I guess how you 0:22:05.920 --> 0:22:09.640 define creativity could could be different, right, But I'm thinking 0:22:09.640 --> 0:22:12.720 of creativity in the sense of sort of agility and 0:22:12.760 --> 0:22:17.440 strategic vision and creative problem solving. I think that's really 0:22:17.520 --> 0:22:20.400 important in the world that we're in right now, being 0:22:20.520 --> 0:22:24.960 able to creatively problem solve with new issues that are 0:22:26.000 --> 0:22:27.880 rising sort of every day. 0:22:28.520 --> 0:22:30.080 And so how do you see the role of chief 0:22:30.119 --> 0:22:33.760 privacy officer evolving in the future as AI technology continues 0:22:33.800 --> 0:22:36.680 to advance, Like, what step should CPOs take to stay 0:22:36.680 --> 0:22:38.600 ahead of all these changes that are come in their way. 0:22:39.880 --> 0:22:43.800 So the role is evolving in most companies, I would 0:22:43.840 --> 0:22:48.800 say pretty rapidly. Many companies are looking to chief privacy 0:22:48.840 --> 0:22:52.360 officers who are ready understand the data that's being used 0:22:52.359 --> 0:22:55.840 in the organization and have programs to ensure compliance with 0:22:56.000 --> 0:22:59.720 laws that require you to manage that data in accordance 0:22:59.760 --> 0:23:01.879 with day to protection laws and the like. It's a 0:23:02.000 --> 0:23:08.280 natural place and position for AI responsibility, and so I 0:23:08.280 --> 0:23:11.040 think what's happening to a lot of chief privacy officers 0:23:11.080 --> 0:23:13.760 is they're being asked to take on this AI governance 0:23:13.840 --> 0:23:17.760 responsibility for companies and if not take it on, at 0:23:17.840 --> 0:23:21.600 least play a very key role working with other parts 0:23:21.600 --> 0:23:25.359 of the business in AI governance. So that really is changing. 0:23:25.600 --> 0:23:29.639 And if chief privacy officers are in companies who maybe 0:23:29.680 --> 0:23:33.440 haven't started thinking about AI yet, they should, So I 0:23:33.440 --> 0:23:37.320 would encourage them to look at different resources that are 0:23:37.359 --> 0:23:41.720