Episode 3 - Governing AI, Sustainability, and Responsible Business

00;00;07;23 - 00;00;33;25
Marc Siegel
Welcome to People Behind the Impact, a podcast from Fordham University's Responsible Business Center at the Gabelli School of Business. I'm your host, Marc Siegel. And in this series, we talk with leaders whose work sits at the intersection of business, sustainability reporting, governance, and long-term value creation. In prior episodes, we spoke with Barbara Porco, associate dean of Gabelli and leader of the Responsible Business Center.

00;00;33;27 - 00;01;00;29
Marc Siegel
We also had a great discussion with Jeff Hales of the University of Texas and the International Sustainability Standards Board. So, if you missed either of those episodes, please go back and have a listen today. I am super happy to welcome Andrew Jones, PhD. Andrew is a principal researcher at the Conference Board Governance and Sustainability Center, where he leads research and thought leadership on corporate governance, sustainability and corporate citizenship.

00;01;01;01 - 00;01;31;01
Marc Siegel
His work focuses in particular on AI governance and oversight, sustainability and disclosure trends, board governance and effectiveness, DEI and corporate citizenship, and the wider legal, regulatory and political shifts shaping how companies respond. He regularly advises senior executives in their teams, moderates, and speaks in internal and external events and comments in leading media outlets. Andrew joined the Conference Board in 2022.

00;01;31;03 - 00;01;46;07
Marc Siegel
Prior to this, he was a senior specialist at the world Bank and an assistant professor at the University of Warwick. He holds a PhD in Modern History from the University of Birmingham. Andrew, thanks so much for joining me. It is so great to welcome you to the podcast.

00;01;46;10 - 00;01;51;16
Andrew Jones
Thanks so much, Marc. It's a real pleasure to be here. I'm really looking forward to the discussion today.

00;01;51;19 - 00;02;07;19
Marc Siegel
Awesome. Let's get started. Maybe before we get into AI and sustainability. Can you tell us a little about your career path? I mean, you've got a PhD in modern history, you spent time at the world Bank and as a professor. So how did those prior experiences lead you to the Conference Board?

00;02;07;22 - 00;02;29;03
Andrew Jones
So yeah, I'm delighted to and I as you know, my career paths are always, you know, they always don't always run smooth. Right? I think it's perhaps only in hindsight you can connect the dots and it all sort of looks coherent. But so, I think for me, I've done a lot of stuff, as you said, you've been sort of in, in British academia watching some of the international development system now at the Conference Board.

00;02;29;03 - 00;02;55;09
Andrew Jones
So probably it looks quite nonlinear. But for me, I think there is a through line in all of that, right? Which I think is just kind of probably how in different ways, in different contexts, how institutions are responding to major changes: major social changes, economic changes, political changes, like it's now technological changes as well. So yeah, I started at and in academia and in the UK, that was always really my passion was always research and the humanities.

00;02;55;09 - 00;03;18;06
Andrew Jones
And you just studied history as a, as an undergrad. And my doctoral research, which is at the University of Birmingham, focused on the contemporary history of global humanitarianism and global development, right, and an international NGO. So, I was really interested, I guess, in how that whole sort of how the global aid system and the broader sort of global governance infrastructure sort of evolved.

00;03;18;06 - 00;03;41;27
Andrew Jones
And I thought it became this sort of latter decades of the 20th century. So that was really what I worked on for many years. But I think sort of after my PhD. So, my interests probably became a bit less sort of purely historical, a bit more interdisciplinary. You know, I was very interested in, in climate and development and inequality and trust, you know, none of which really necessarily fit into one discipline.

00;03;41;27 - 00;04;11;10
Andrew Jones
And I became— I think that gradually pulled me more towards sort of applied work. And I left the Academy. I spent a number of years working on sustainable development, which was sort of right after the SDGs were sort of, you know, in that peak moment, where a lot of energy and vibrancy. Right. And I really focused on a few things, but a lot to do with sort of climate, smart infrastructure and urbanization and in sort of emerging markets in Africa and Asia.

00;04;11;10 - 00;04;29;04
Andrew Jones
And that that included a few years at the World Bank, as you said in your introduction, which is really fascinating work, helping sort of get some of these projects off the ground, so that experience all kind of brought me closer to some of these big institutional challenges right now. Like how do you mobilize capital, right? How do you build markets, how do you structure incentives?

00;04;29;08 - 00;04;52;17
Andrew Jones
I think that's probably what ultimately led me to the conference board right there in the Conference Board. For those who listening, you maybe don't know, a very interesting organization, right? We have a long history. We are a business membership organization. We're also a think tank. We sort of sit, I think, at the intersection of research and practice where we're not purely academic, but we're not advocates or consultants as well in the conventional sense.

00;04;52;17 - 00;05;18;03
Andrew Jones
I think the what we really do is we help senior business leaders across all kinds of functions, understand complex issues. And yet within GSC, I mean, the Governance and Sustainability Center. So, as the name suggests, focus on ESG, corporate governance, sustainability, citizenship, which I think became a kind of natural, at least I hope became a natural home for the kind of work I've been doing and still in a different kind of way now, much more of a big business lens.

00;05;18;03 - 00;05;31;10
Andrew Jones
But I hope he's still asking some of the same questions I started out with. But, you know, it's always so it's interesting to reflect on your trajectory, right, Marc? And maybe I've imposed that coherence on it there that it didn't have when I was, when I was working my way through it.

00;05;31;13 - 00;05;57;06
Marc Siegel
Yeah, I agree, like, none of us suffer from the linear type of, career trajectory. It's much more serpentine, as you pointed out. You know, that interdisciplinary type of thoughts and work definitely resonates with me. And you seem to work at the intersection of governance and sustainability and corporate citizenship and now AI as well. And those kind of appear, at first glance, to be separate fields.

00;05;57;12 - 00;05;59;07
Marc Siegel
But how do you see them fitting all together?

00;05;59;09 - 00;06;16;17
Andrew Jones
Yeah, it's a great question Marc, because within our part of the Conference Board, they sort of they are separate worlds, right, and they have their own audiences and their own leaders and their own cultures, but they also do fit together in some— or at least overlap and intersect in some quite interesting ways. Right. So, I think for me, I probably see them in very different ways, right.

00;06;16;17 - 00;06;44;23
Andrew Jones
Connected by some of these core questions, right. Around how do companies govern I guess, their responsibilities or their risks in a complex environment, I guess an environment that's becoming more complex as well. Right. So, whether sustainability, I guess, asking how companies manage environmental risks or corporate citizenship, asking how companies, I guess, contribute to communities and societies and manage some of those social risks or obviously governance, who are very much focused on boards of management, sort of overseeing all of that credibly.

00;06;44;23 - 00;07;04;07
Andrew Jones
And I think AI fits into that and overlaps and layers and in interesting ways, because I think it's, as we know all see, it's no longer just if it ever was just a technology issue, I think it's raising questions about accountability and risk and environmental and social impacts across the board. Right. So yeah, I think those do all fit together in an interesting way.

00;07;04;07 - 00;07;20;19
Andrew Jones
And I think, interestingly, the language of ESG might have given them some sort of coherence as well as an umbrella. I think we've moved away from our language, but I still think they converge and in interesting ways. I don't want to overstate as well. I think interesting convergence for sure.

00;07;20;21 - 00;07;38;26
Marc Siegel
Yeah. I think there's so much you said there that we'll try to unpack as we go through this and address some of those different specific topics that you mentioned. But I had a just a slightly different question. As you were mentioning all those items, it occurs to me that you and the Conference Board generally sits sort of adjacent to the C-suite.

00;07;38;28 - 00;07;53;27
Marc Siegel
And some of the topics in particular that you're focused on in the Governance and Sustainability Center can be pretty polarizing right now. As you alluded to, how do you approach topics like sustainability, DEI, climate and AI, where the public debate can be pretty polarizing?

00;07;53;29 - 00;08;14;10
Andrew Jones
That's a great question, Marc. And I think, to be candid, this is a genuine challenge, right? You know, I think I think it is I think those topics you just mentioned right there, business issues, but they do sit inside or are shaped by, you know, broader political debates. But I think cultural and social debates as well. And obviously, part of the reason they've become polarizing, is one we live in a polarized society.

00;08;14;10 - 00;08;37;09
Andrew Jones
But two, they do touch some of these big political questions, right, about where it's about the role of business in society. Right. And how should we even, should business at all be responding to whether it's climate risk or inequality or technological disruption? These kind of questions and never reacts attract scrutiny from all kinds of stakeholders, right? Including activists and political officials and the courts.

00;08;37;09 - 00;08;58;16
Andrew Jones
Right. So, it is a challenging I think there is a political milieu around all this. Right? I think for us, we take pride in being nonpartisan and being nonpolitical in and trying to be dispassionate. And I'm really focused on, I think, evidence and data. And I think we always should aspire to try avoid as much as we can, you know, importing some of that debate into the corporate setting.

00;08;58;16 - 00;09;26;18
Andrew Jones
Right. And I think for us, our members and the wider private sector needs practical analysis, right? It doesn't necessarily need slogans, but it is a challenge. And, you know, I'm sure, as you seem to mark, right, there's been a lot of backlash in the last few years. And whereas backlash against ESG or backlash against certain aspects of E&S, and it can make the work difficult, I think we just have to sort of understand it, separate the signal from the noise and respond in a disciplined way as much as we can.

00;09;26;21 - 00;09;44;14
Marc Siegel
I think that that's right. I mean, I think, you know, you and I have talked about this offline, just trying to adjust language and changing vocabulary. And what you've been doing is giving voice to help companies navigate through this as well, because they're seeing the same challenges that you and I are seeing. Right. And they're trying to figure out how to do this.

00;09;44;14 - 00;10;09;13
Marc Siegel
So, I think it's extremely important work that you've been doing and I admire it. Let's talk a little bit now about the sustainability reporting in responsible— what I call responsible reporting. And you've done work on sustainability and disclosure trends. And I've attended 1 or 2 of your excellent roundtables on this. From your vantage point, where are companies right now on sustainability disclosures or is anybody advancing them or are they pulling back?

00;10;09;13 - 00;10;16;03
Marc Siegel
Are they recalibrating or all of the above? And I guess along the continuum where a companies between greenwashing and green hushing?

00;10;16;09 - 00;10;36;16
Andrew Jones
Yeah, this is a great question, Marc. This is a big topic, right. And that’s a topic you have a lot of expertise and wisdom in. And we should say you mentioned our roundtables. You've been a fantastic contributor at some of those roundtables. So, thank you very much. I think for us and I should— we have some benchmarking capabilities of disclosure that I think we've developed at the conference.

00;10;36;16 - 00;11;01;05
Andrew Jones
Well, they're quite sophisticated. You know, we're looking at big, big groups of US public companies. I think it is probably all of the above things you mentioned. Right. Kind of at the same time. But recalibrating is probably, I think, a fair assessment. Right. It I guess to be candid, there has been some pullback. Right? I mean, and to give you, give you some hard data where we through our benchmarking capabilities, we were able to track last year the number of Russell 3000 companies.

00;11;01;05 - 00;11;23;24
Andrew Jones
Right. So, kind of a big slice of corporate America. That put out an annual sustainability report compared to the year before. And actually, that number fell by about 20%. Right. It's about a fifth. So, some contracting of disclosure I think that is it's not a majority, but it is significant. But it's interesting, when we dug a little deeper into that data, that fall was really concentrated amongst sort of smaller and mid-cap firms.

00;11;23;24 - 00;11;58;05
Andrew Jones
Right. We actually saw very little shift in larger cap firms. Right. Sustainability reporting has become routine. I think there's significant depth to it. Basically, virtually all large public companies are disclosing on emissions and climate risk, for example. So, I think the real story there is perhaps not that sustainability reporting is fading or being eroded, but I think it's perhaps that it's been institutionalized that the larger companies, while perhaps some of the smaller firms, are perhaps more likely to reassess, you know, what's the cost, the risk, the value of such reporting in, to be frank, are less conducive political and legal environment.

00;11;58;05 - 00;12;29;14
Andrew Jones
So, it's an interesting moment in sustainability reporting. I think it's I think more broadly, we're probably seeing a shift as well, from quite expansive, aspirational ESG storytelling to something that I think is maybe a bit more controlled, evidence based, defensible, legally reviewed. Right. So, I think that that's and that probably reflects a broader trend in the whole sustainability space, if I'm honest, is as we as it matures and as some of the assumptions, that change about where it's about policy, where about technology, etc.

00;12;29;16 - 00;12;50;18
Marc Siegel
Yeah, it was fascinating following your work over time, over 18 months or so, and you were very quick to notice that companies were at first just slowing down their ESG reporting and almost waiting for what the rest of the market is doing so that they could benchmark as they go. And then deciding, okay, well we're going to be reporting something.

00;12;50;18 - 00;13;08;08
Marc Siegel
Now, let's figure out the content that goes in it. You know what I asked you was sort of a as of a point in time where are they now? But it's also, I would encourage people to follow your work so that you could see over time how those trends are developing. And then you mentioned climate. So, let's drill down for a second on climate.

00;13;08;08 - 00;13;25;27
Marc Siegel
And you've looked specifically at climate disclosures across the Russell 3000 S&P and the GHG reporting targets regulatory preparedness board oversight. What are companies getting better at with respect to climate disclosures specifically and where are the persistent weak spots for fighting?

00;13;25;28 - 00;13;46;27
Andrew Jones
Climate disclosures is a fascinating area, right. It's probably one of the areas I really find interesting. And you can sort of endlessly drill down. Right. And there's more depth to it. I would say that's talking in the broad sense, I think companies as a whole of the larger companies’ multinationals have got, I think, got much better at measuring and disclosing greenhouse gas emissions.

00;13;46;27 - 00;14;07;15
Andrew Jones
Right. Basically, scope one and scope two emissions virtually, or like to use it, for example, the S&P 500 right thing like over 90% of S&P 500 firms are reporting on scope one and scope two greenhouse gas emissions. Right. And they're increasingly starting to use external assurance and auditing to verify those numbers and put some more sort of financial muscle behind it.

00;14;07;17 - 00;14;33;19
Andrew Jones
Many are disclosing on board oversight of climate, many a connecting climate to enterprise risk management. So, and as, you know, Mark, a big driver a lot of this is sort of preparing or gearing up for new regulations. I think underpinning all of that, Marc, is probably a much better set of data architecture writing. Companies have developed these good data systems for climate, more formal governance, clearer ownership, etc. so I think that's a really positive change.

00;14;33;19 - 00;14;50;01
Andrew Jones
I think where perhaps you have about like weak spots, I think a persistent weak spot probably, which a lot of the audience will recognize is scope three emissions. Right. So, and I've said emissions in the wider value chain, which for many companies is actually the vast majority of their footprint. Right. I think that's not surprising. Right. I think we all know it.

00;14;50;01 - 00;15;09;25
Andrew Jones
It's super hard, right? You know, it's sort of it these are emissions that are on in the company’s control. It could be supplier’s product use. Right. Investments if you're a financial firm and these often sit with different suppliers, different third parties. Right. Could be in different countries. And it is hard to get that data right.

00;15;09;25 - 00;15;31;19
Andrew Jones
I think because of that, a lot of scope free data that we see at least of what's reported still relies on estimates and assumptions. You know, those assumptions can change over time. But I would I would say companies are improving in that space, too. And I think companies have been doing a lot, whether it's building supplier engagement or updating the procurement and so on, I think, and also gearing up for regulations that will require scope free disclosure.

00;15;31;19 - 00;15;48;07
Andrew Jones
So, it's still a work in progress on some of these more complex climate metrics. But as a whole, I think climate reporting is quite mature and quite robust. And make the next the next frontier is probably sort of bringing that extra level of order to auditability and assurance to it all.

00;15;48;09 - 00;16;10;09
Marc Siegel
Super helpful context. Just for any listeners who may not be as familiar with greenhouse gas GHG protocol, scope one are the emissions that the company themselves are putting out into the atmosphere. Scope two are the inferred emissions for the energy that the company purchases and uses in its environment. And scope three is kind of everything else, right?

00;16;10;09 - 00;16;37;20
Marc Siegel
It's from the beginning of time, the beginning of the lifecycle and the suppliers all the way through and of use and the customers. That's all captured in scope three. And so that's what makes it so hard as you mentioned. Also, it probably should be noted that right now, the GHG protocol, which is the organization that put out scope one, two and three definitions and guidance, is actually rethinking some of the aspects.

00;16;37;23 - 00;17;01;05
Marc Siegel
And so we'll see how that all falls out, whether it gets easier, more difficult or both. Another question for you and this is more personal for me, I like to describe my role at the Responsible Business Center as promoting responsible reporting. And I think about all forms of corporate communications to investors. When I think about that, whether it's financial sustainability or some other communication channel.

00;17;01;05 - 00;17;07;29
Marc Siegel
And I guess I'm wondering, does that resonate with you and how do you think about the different forms and channels of corporate communication?

00;17;08;01 - 00;17;23;10
Andrew Jones
It does resonate, Marc. And I think it's it's a really helpful and really smart way, I think, to to look at it in a really helpful lens because and It does resonate, Mark. And I think it’s a really helpful and really smart way, I think, to look at it in a really helpful lens because— and we've actually had a lot of discussions really, where I think, as we all know. Right, corporate reporting is how to put it right. It is fragmented across many channels. Right. actually had a lot of discussions really, where I think, as we all know. Right, corporate reporting is how to put it right. It is fragmented across many channels. Right.

00;17;23;10 - 00;17;47;21
Andrew Jones
And some of the some are very obvious, like the financial filings or the SEC reports, sustainability reports, proxy statements, etc., but also websites, press releases, social media. Right. And I think obviously a lot of these channels are owned by different teams right now. Legal might be involved somewhere. Sustainability on the sustainability report. Investor relations. Right. But I think if you are an external audience, you don't necessarily experience it that way.

00;17;47;21 - 00;18;11;00
Andrew Jones
Right? They're all just the corporate voice. And I think what we're been tracking a lot recently, I think what's really interesting about is this perhaps matters now more than ever, because with the help of AI and other technologies, it's actually now a lot easier to assemble a company's body of communications. Right? I think we do live in a world where it's like investors or activists or plaintiffs, right, can quickly compare language across the years.

00;18;11;00 - 00;18;31;26
Andrew Jones
Right. And look at, oh, is there inconsistencies between what was said in that annual report, what was said in that ESG report, which is probably a new trend, or at least an intensified trend. Right. So, I would argue and I hopefully I this you agree, given everything you're involved in, that there are that responsible reporting you say, probably matters more now than ever.

00;18;31;26 - 00;18;49;21
Andrew Jones
Right. And the big question is about how do you govern communications, right. How do you have consistency? How do you have contacts? How do you have discipline across channels, right. Big, big, big challenge. And I only see that getting sharper as AI develops some more and more people are harnessing it for this kind of purpose.

00;18;49;24 - 00;19;11;00
Marc Siegel
Yeah. In a prior life, I used to manage a team of 20-25 analysts that would have to dig and try to read all these different forms of communication, could find inconsistencies and see if the story being told by management teams in their earnings calls and on their website actually matched up to the story being told by the regulatory filings.

00;19;11;02 - 00;19;40;06
Marc Siegel
And that was a much less of a technology push. It was much more of a human and labor-intensive type of work that had to be done to see that. And now, as you're mentioning some of those things, AI can help you evaluate and, some still you need some humans in the loop to figure it out, but it is an evolving type of thing, and it makes it all the more important for, in my mind, companies to not think of these different communication channels as siloed.

00;19;40;09 - 00;19;58;07
Marc Siegel
And to that end, I'm wondering if you're seeing over years, probably over the last number of years, are you seeing sustainability reporting becoming a little bit more like financial reporting with stronger systems, controls, assurance, expectations or their limits to the analogy to financial reporting?

00;19;58;09 - 00;20;19;07
Andrew Jones
Great question. I think. Yes, seeing it becoming more like financial reporting in some really important ways, but probably, you know, don't want to overstate the analogy, but I think, yes, I think when I look at sort of the manner in which sustainability reporting is becoming more regulated, you know, and this is, as you know, there's a whole patchwork of regulations out their market.

00;20;19;07 - 00;20;39;18
Andrew Jones
Some companies are subject to more than others, I think, more controlled, more data driven. Right. I think more subject to assurance. We have a lot of, I think, interesting data that shows more and more companies every year disclosing the use of external and assurance to verify their, for example, their emissions data and so on. And I think behind the scenes, I see companies are then to do that.

00;20;39;18 - 00;21;00;03
Andrew Jones
They're building processes that do look a bit like financial reporting, right? When it comes to ownership structures and internal controls and sign off procedures and so on. And regulation, I think, is such a key driver of this. I just wouldn't push it all the way because I think I think as we kind of spoke about earlier, maybe we're talking about scope three emissions and sustainability data still depends on some estimates.

00;21;00;03 - 00;21;27;14
Andrew Jones
Right. And then the methodologies are evolving and there is perhaps some subjective judgments in there. But definitely is a process is underway, I think, you know, and one of the dimensions of that I think is also a closer coming together of sustainability and finance. And we definitely hear sometimes anecdotally that there are a lot closer relationships between the sustainability function and the finance function than we might have seen in prior years for the kind of reasons we spoke about.

00;21;27;14 - 00;21;35;26
Andrew Jones
So, yeah, it's interesting watching this evolve in, in real time, and it's interesting to reflect on where it will end up and how far it will go.

00;21;35;29 - 00;22;07;08
Marc Siegel
Yeah, your history background makes me think that maybe you'll appreciate this thought process, at least in this country. The SEC promulgated Securities Act of 1933 and then 1934, which really is the foundation of modern financial reporting and supplemented by Sarbanes-Oxley about 20 to 25 years ago. And so, there's been, say, 90 years of maturation of financial reporting, right? Building up to all the infrastructure that you have to have today to do your financial reporting, whether it's quarterly or semiannually.

00;22;07;08 - 00;22;34;20
Marc Siegel
And that's a whole other topic. But when you think about sustainability reporting as compared to the 90 years is probably, what, 90 months or, in some cases, 90 weeks of actual infrastructure and maturation in the sustainability reporting. So, we are seeing a I agree with you. I think we're seeing at least some migration towards some of the controls and things that you might see in financial reporting.

00;22;34;23 - 00;22;54;10
Marc Siegel
When I first started doing this, 6 or 7, 8 years ago, I would routinely hear about huge companies that had big sustainability reports, and their levels of controls were an email from one person to another around the world saying, where'd you get this number? Oh, Vinnie told me, perfect, check. We don't get that as much anymore. And certainly, that won't survive any assurance.

00;22;54;10 - 00;22;55;02
Marc Siegel
Expectations.

00;22;55;07 - 00;23;09;17
Andrew Jones
No, no. And I just sort of I didn't see that. So, it's wonderful to put it in that longer history. I think it does also underscore that this is a new, relatively new discipline as well. Right. And I think that's only going to continue to go. As you were talking, I was thinking about some of that the target setting.

00;23;09;17 - 00;23;31;25
Andrew Jones
Right. You know, I think I think a lot of companies have big climate targets. Right. Would be net zero by X year. Right. A lot of those were set not necessarily with much sort of data behind them. Right. There were perhaps aspirational targets. So that'd be figured out. I think now we are seeing a lot more like linkage of sort of that target setting to like capital allocation and financial planning and breaking down.

00;23;31;25 - 00;23;48;16
Andrew Jones
And I think that's a really positive trend. Right? I think it's much less glamorous and it's more it's more sort of the hard nuts and bolts. But I think that is a maturing. And I think we'll continue to see that happening in different ways within the distribution. It's a positive trend for sure.

00;23;48;18 - 00;24;15;18
Marc Siegel
So, let's go from maturing and maybe the teenage years of sustainability reporting to something that is very immature and nascent and just in the last couple of years, and that's AI. So, let's talk a little bit about artificial intelligence. And I really loved your recent report on AI governance from what's called From Principles to Practice governing AI in the Corporation, because it looks at how large companies are putting AI governance into practice.

00;24;15;18 - 00;24;23;12
Marc Siegel
And so, what prompted that research? Now what were you seeing that sort of spoke to you that said, we really need to put something together on this?

00;24;23;14 - 00;24;42;23
Andrew Jones
Yeah, sure. I mean, we can't have a podcast discussion of that talking about AI, right? Marc. It just wouldn't be right. But we have to because how could you avoid it? And I think, you know, we started to about how I'm in the— you know, I sit within the Governance and Sustainability Center at the conference Board. I think I has been sort of moving up our agenda for, for the last few years.

00;24;42;23 - 00;25;06;13
Andrew Jones
Right. I think guys really in tandem with it moving up the business and policy agenda. Right. And so quickly I moved, I think, from being, you know, sort of a new shiny thing to actually starting to be integrated into the business. And for our members, you know, the companies we work with, whether they're corporate governance leaders, sustainability leaders or even citizenship leaders, this has become— the appetite here is huge.

00;25;06;13 - 00;25;24;26
Andrew Jones
Right. To really understand sort of how do we how do we build this right. How do we governance AI at scale? How do we manage risks? How do we oversee AI use how do we understand and do something about AI's broader societal impacts? Right. These are all the kind of big questions we're grappling with. I think I think recognizing that we've been we've been active for a while.

00;25;24;27 - 00;25;45;29
Andrew Jones
I, we've been convening executives in off the record discussions, we’ve been doing disclosure analysis. And we did recently and this this led to the piece you mentioned we did at the start of the year, a survey of 130 executives across kind of ESG executives, really across governance, sustainability and CSR as well. So that was really the impetus.

00;25;45;29 - 00;26;05;09
Andrew Jones
Right. And I think we've recognized that companies have been facing these familiar governance problems. And the research really, we try to get into the surface of these issues a little bit, really understand where companies are, where they're heading, what comes next, you know, and then it's such a fast-moving space that, what comes next is, is suddenly what's just happened.

00;26;05;09 - 00;26;11;21
Andrew Jones
But I think hopefully we've brought, some practical insights to, to a very complex space.

00;26;11;23 - 00;26;22;11
Marc Siegel
I think you did. And that leads to the next question, which is a lot of companies have AI principles, but, you know, report focuses on moving from those principles to, actually practice. And so, what does that shift actually look like inside a company.

00;26;22;19 - 00;26;38;26
Andrew Jones
Yeah. So, part of the framing was I think that just as recently as about a year ago, it felt like there was such an emphasis on developing and articulating principles right there that we will be fair, we will be transparent, we will be accountable, we will ensure human oversight, security, etc. and those principles really mattered.

00;26;38;26 - 00;26;58;18
Andrew Jones
I think that's a really important step the companies have to go through right in their AI governance journey. But they're also clearly only the starting point. Right. And I think the harder question then becomes sort of what are those mean? Right. How do you translate those? What do they mean inside a company? Right. On an average day when someone wants to deploy an AI tool or someone wants to use particular data sets, right.

00;26;58;20 - 00;27;16;11
Andrew Jones
So, I think we are in the midst of this transition at many companies from principles to practice. Right. And I think that means different things to different companies or at different companies are obviously at different stages of their AI journey. They have different kind of relationships with AI. But I think, talk about some sort of broader learnings.

00;27;16;11 - 00;27;36;21
Andrew Jones
I think the first step is visibility, right? I think companies need an inventory of AI use cases because you can't govern what you can't see. That includes those tools are being developed internally as well of what you're procuring or what's already in the existing software and so on. Then I think the next step is risk and risk tearing.

00;27;36;21 - 00;27;55;24
Andrew Jones
And I think AI risk is such a fascinating area. Right. And clearly not every AI use case needs the same level of scrutiny. Right. You know, I think [inaudible] an internal tool you're using to draft reports is obviously very different from something you're using and hiring or lending your healthcare. Right? So, I think it's sort of understanding that risk matrix and how you appropriately govern.

00;27;55;27 - 00;28;19;20
Andrew Jones
And then I think after that it's accountability, right. Who actually owns this within the organization. Right. Who reviews data, who decides when human oversight is required. And suddenly all these steps sound quite simple when I say them. Right. But they're actually they're actually just so extremely complex, particularly in large global organizations, because there's so much AI is moving so fast and there's so much of them, we can't fully see.

00;28;19;23 - 00;28;27;29
Andrew Jones
And so much is still up for debate, I think, which is making for a really interesting time and also keeping a lot of governance leaders up at night, for sure.

00;28;28;01 - 00;28;47;12
Marc Siegel
Yeah, I think I want to get into the risks in a minute. But as you were talking, I also heard you say accountability and where this stuff is owned. And I think I want to get a little more of a double click on that. Right. Because where does AI governance belong organizationally? Is it a board issue? Is it a management committee issue?

00;28;47;14 - 00;28;54;26
Marc Siegel
Legal compliance, IT, cybersecurity, all of the above. And maybe this is part of the maturation conversation.

00;28;54;27 - 00;29;11;20
Andrew Jones
Yeah, I think it's probably all of the above. Right. But playing out in a lot of different conflicts ways. Right. And that doesn't necessarily mean everyone owns it equally. Right. Or that therefore because it's everywhere, it's vague. And who knows, right. Who actually makes the decision? I think there is a lot of [inaudible] systems. You got to start the board.

00;29;11;22 - 00;29;30;09
Andrew Jones
The board has to oversee AI as a strategic risk and governance issue. Right. That doesn't mean that managing every individual thing that's AI related, right? But they have to have some level of oversight and they need to be asking the right questions. Right. And then that's sort of the key, C-suite management level. Clearly you need some kind of enterprise level structure, right.

00;29;30;09 - 00;29;47;24
Andrew Jones
And that's going to look differently at different firms that we've definitely seen in some of our data. And some of our convening is a trend towards having some kind of cross-functional AI group, right? Whatever you call it. Right. The AI Council or the AI Governance Committee. And in the some of these people who need to be sat on that, right, you know, like technology.

00;29;47;26 - 00;30;07;19
Andrew Jones
Yeah. That how you cannot have the tech, the tech guys in the legal and compliance. Right. Have to be there. IT and cybersecurity, they all seem very obvious. Right. But I would argue that you need a bunch of other people to. Right. And how can you not have HR in there? You know, given how AI affect hiring and workers planning, you can make a strong argument.

00;30;07;19 - 00;30;29;16
Andrew Jones
As much as I think not many companies are doing it yet for sustainability to have a strong voice in that group, given some of the environmental and social implications. Procurement, I think because of many of these risks are actually coming from third party vendors. So that then in turn raises the question, right? Yeah. What point do you just have everyone from the organization involved but clearly need to and companies need to manage and avoid extremes.

00;30;29;16 - 00;30;45;20
Andrew Jones
Right. You clearly can't treat AI is just an IT issue, but you also need to avoid creating this sprawling governance process that slows everything down. So, I think it's hiring the right people in the room, having the right structures, hiring the right vehicles, and then having at the top of the organization clear oversight of strategy.

00;30;45;22 - 00;31;17;20
Marc Siegel
Sounds very reminiscent in many ways to the ESG discussion we were having a couple of years ago of where does ESG fit in the organization as well? And there is a board committee is that across the board? Is it management and how? But the things that you're saying makes me think that in any of these emerging areas, making sure you've got a cross-functional, interdisciplinary perspectives involved in all these things is going to be super helpful because none of them reside particularly well in only one organization, one part of the organization.

00;31;17;22 - 00;31;39;17
Marc Siegel
So, let's get into the AI risks. So, what are the biggest AI risks that companies are recognizing right now. I think of a million of them. Right. Cybersecurity regulatory exposure model bias, intellectual property reputation, workforce disruption, data privacy, environmental impact. Like how do you get your hands on around all of this, right?

00;31;39;17 - 00;31;56;13
Andrew Jones
I mean, that list just goes on and on, right? And it's such an it— I actually think is such an interesting our discussion right now just about sort of how to even come up with like an AI risk taxonomy. Right. And, and to what extent is AI risk a separate category, and to what extent is it just amplifying some of those other categories that you just said?

00;31;56;13 - 00;32;10;21
Andrew Jones
Right, Marc, I think that's a debate that's probably gonna going to continue on, I think for a while, as companies still figure out some of the processes, I'd say from a corporate objective. And, you know, we've seen this in some of our benchmarking data, I'd say companies are probably right now they're seeing a lot of these risks.

00;32;10;21 - 00;32;33;12
Andrew Jones
Right. And they're the ones are prioritizing. And I think the kind of immediate ones, right, that the immediate enterprise research such as cybersecurity, data, regulatory compliance, etc. Right. And actually, in the survey we mentioned, cybersecurity was by some margin, actually the most consistently identified top AI risk by the audiences that we spoke to. And I think that makes a lot of sense.

00;32;33;12 - 00;33;00;17
Andrew Jones
Right? I think it's a media it's familiar. It's already sort of there. You know, we've been a lot of focus on cybersecurity in the last few years. And I think a lot of people are starting to get quite concerned. And about the way in which AI, from a cyber perspective, of potentially expands the attack surface, right, can maybe make fraud and phishing and malicious code a lot easier to generate, and therefore a lot easier to, to attack companies.

00;33;00;17 - 00;33;18;00
Andrew Jones
Right. And a lot of I think there's a lot of a lot of anxiety in that space. Right. So, cybersecurity right now a big, big area of focus. But seeing other areas rising up too, I think privacy and data protection consistent issue. Given that focus on data. And how do you ever truly know what data is being entered into systems.

00;33;18;00 - 00;33;45;16
Andrew Jones
Right. Regulatory compliance is moving up fast. I think as, as the fragmented regulatory landscape continues to take shape both in the US and globally. But there's less commonly recognized risk to including some of the ones you mentioned right around workforce, around environmental risks, around social. And while they may not be fully on the agenda yet, I think they will continue to capture attention and gradually make their way into these corporate risk discussions.

00;33;45;19 - 00;33;50;21
Marc Siegel
Any view on how companies are disclosing these kinds of risks in the corporate reporting these days?

00;33;50;21 - 00;34;06;09
Andrew Jones
Yeah, yeah, we actually we did a big project on this last year that for me, it was a really interesting piece of work. Right. Were quite, quite a simple project. We simply looked at, the S&P 500. Right, a big large cap companies. And we looked at what they're saying in the risk factors section of their annual report.

00;34;06;09 - 00;34;25;25
Andrew Jones
Right. So, for those who don't know, this is this is, you know, common standardized practice for companies to disclose their material risks in their annual 10-K SEC report, along with potentially some ways in which they mitigate. Right, to some extent. That's a lot of its legal boilerplate, but I think it is an interesting perspective on what companies are saying and how they think about risks.

00;34;25;25 - 00;34;48;00
Andrew Jones
Right. Let's give you, I think, a really interesting stat. As recently as three years ago, it was about 12% of S&P 500 companies had an AI related risk in that section. Last year's 83%, right. So, I very, you know, that's in the space of a few years AI risk, I guess, has become mainstream. And I think what you're not going to start seeing is companies with several AI related risks in that section.

00;34;48;00 - 00;35;12;06
Andrew Jones
Right. Talking whether it's about cybersecurity, about regulations, about workforce, about privacy and so on. So, companies are definitely starting to say a lot more about AI risk, at least in certain places. Right. And in the next stage is probably just going to be continue to get more specific in those, as well as have to be more specific on how to governing them and how they're mitigating them.

00;35;12;06 - 00;35;18;03
Andrew Jones
Right. But I risk here to stay as a top, top corporate concern.

00;35;18;05 - 00;35;39;17
Marc Siegel
Super helpful context. And now I wanted to touch on some of that intersection you were alluding to between AI and sustainability and environmental impact. Right. So, let's get into that a little bit. You've written about AI and environmental sustainability, and I kind of like the distinction you make between sustainable AI and AI for sustainability. And can you walk our listeners through those two lenses.

00;35;39;24 - 00;35;58;15
Andrew Jones
Yeah, sure. So, the background here is that we spoke earlier about some of the work we've been doing in the Governance and Sustainability Center on AI, and we did have some specific work on the intersection of AI and sustainability. And we kind of came at that from two different, I guess, two different fronts as much as they overlap, one of which is sustainable AI.

00;35;58;15 - 00;36;16;24
Andrew Jones
Right. So actually, looking at the environmental footprint of AI. Right. And really that typically means the data center build out, right, which we think we all know is sort of scaling more and more every day. And the electricity use, the water use, as well as some of the more indirect stuff like the, you know, the electronic waste and hardware and so on.

00;36;16;26 - 00;36;33;28
Andrew Jones
So, I think that's becoming and that is interesting in watching that happen in real time, how that seems to be becoming more and more of a major concern from a political aspect, from a business aspect, every day, really. And there's a whole depth of analysis there. On how do you, you know, how do you potentially try and mitigate some of these demands at the same time?

00;36;33;28 - 00;36;59;09
Andrew Jones
Now also looking at AI for sustainability. Right. And we look at these as two lenses. Right. Which is at the same time I guess the AI scaling is having an environmental impact. AI itself can be used for to some extent to potentially improve environmental outcomes right now, whether that's deploying AI to be more efficient in energy use or optimizing buildings, climate risk modeling, product design, etcetera, etcetera, etcetera.

00;36;59;09 - 00;37;23;11
Andrew Jones
So, sort of interesting tension there, right? Because I think the tension is both of these could be true at the same time. Right? I on the one hand can be increasing environmental pressure and at the same time potentially helping companies do something about some of their environmental challenges, whether it's waste emissions and efficiency, etc. So, I guess when we promote this view, I guess we're trying to sort of not say that AI is inherently good or bad for sustainability.

00;37;23;11 - 00;37;43;26
Andrew Jones
I think it's trying to look at the specific use cases and where we can use AI in a way to create environmental value that justifies some of that footprint. Right. Yeah. And that's going to look different at different companies. Right. You know, for tech companies, obviously this is a highly material question because they're building the infrastructure for perhaps the broader universe of enterprise users.

00;37;43;26 - 00;38;03;27
Andrew Jones
The footprint might be smaller directly, but it still matters indirectly because they're still may the ones who are using these tools. Right. And they're there, and therefore it's part of their scope free to go back to that concept we introduced earlier. So, it's a really, it's a really complex area, but it's a really fascinating area. And I only expect attention to grow here from a variety of angles.

00;38;04;00 - 00;38;23;04
Marc Siegel
I just think that's such a helpful distinction and framing so that it doesn't lead to the polarization. We were talking about some earlier. Right. And just being able to sort of describe this as well: There are some negatives, there are some positives, and it really depends on how companies are using it. And depending on how they're deploying it.

00;38;23;06 - 00;38;43;05
Marc Siegel
We could try to look toward models that minimize the risks and the negative impacts and maximize the positive impacts. I think that's super helpful framing and just having a great impact on the conversation overall. And we got to get into data centers and you mentioned them. Data centers are getting all the attention because of the power and water demand.

00;38;43;10 - 00;38;52;01
Marc Siegel
So how should companies think about the environmental footprint of AI, especially if they're not even technology companies, but they are increasingly using those kinds of AI tools?

00;38;52;04 - 00;39;12;27
Andrew Jones
Yeah. I mean, you say, right, we can't avoid this, can we? I think what's so interesting about the data center story, right. It's not just an infrastructure story anymore. Right? It's a political flashpoint. Right? It's an environmental flashpoint in particular, when we talk about power, demand and water use and local community impacts. Right. I think it's still evolving, but clearly is becoming quite complex.

00;39;12;27 - 00;39;30;11
Andrew Jones
And I think it's becoming more and more of a political issue every day. And as you said, for the for the tech companies, right, for the for the so-called hyperscales, right, who are building out and or financing or driving this bill. Right. This is obviously a material issue directly. Right? It's their footprint. Right? I should be out of the tech, I'm guessing, [inaudible].

00;39;30;11 - 00;39;45;10
Andrew Jones
I think we're quite interesting ways to try and mitigate some of this where it's through, if you know, some of the water efficiency aspects or even building their own off grid energy and so on. Right. So, so this is a you know, it's not a simplistic debate and there's a lot going on for the non tech companies, which is probably most companies.

00;39;45;10 - 00;40;02;13
Andrew Jones
Right. They're not building data centers. Right. But they are I guess customers of data centers when they're using you know their frontier generative AI models. It's complex right. And I think companies need to sustainability is need to tread a bit carefully here. Right. Because on the one hand they need to avoid just treating this is irrelevant.

00;40;02;18 - 00;40;21;22
Andrew Jones
Right. You know, it's not a problem. It's not our data centers, whatever. But also, we need to avoid overstating it, you know, as a material climate issue without the evidence. Right. And I think for most of these companies, that footprint is going to show up in, it's going to show up in their procurement, in their scope three emissions, in their tech strategy and their data.

00;40;21;22 - 00;40;43;09
Andrew Jones
So, I think the non-tech companies need to think through kind of what tools are they using? For what purpose, through which vendors. Right. And what is their footprint. And then maybe ask some of these cloud providers and AI. Then there's some harder questions, maybe about what energy is being used, how is water being managed efficient at the data centers and ask for more data, right.

00;40;43;09 - 00;41;05;01
Andrew Jones
And ask for more disclosure data. I think that will help perhaps make it more visible just how significant the AI related environmental footprint is. But it's still a complex area. And definitely anecdotally, we hear from a lot of sustainability leaders still unsure how to approach this issue. Right, and how to frame it in comparison to some of their broader, environmental challenges.

00;41;05;03 - 00;41;09;12
Andrew Jones
Right. Whatever it's around decarbonization and so on. So, it's still a I'm still a work in progress.

00;41;09;12 - 00;41;25;26
Marc Siegel
I think I could see that. And I think that makes a lot of sense given where we're at. And it is difficult and you might not have a whole lot of data, you might just have intuition as to so how it's falling out and how it's impacting. But you might not actually have hard data to support a conclusion one way or the other.

00;41;25;28 - 00;41;38;06
Marc Siegel
On the opportunity side, I imagine and hope you've seen some anecdotes of where AI is actually helping companies improve sustainability performance and not just reporting, but maybe operational outcomes.

00;41;38;08 - 00;41;54;19
Andrew Jones
Yeah, it does. And actually, we have some, I think really interesting data on this where in the same survey we spoke about earlier where we had a section that was just for sustainability heads, it was quite interesting because we asked two questions, right. The first one we asked is where are you currently using AI for sustainability related stuff?

00;41;54;19 - 00;42;15;02
Andrew Jones
Right. And we also asked, where do you see the biggest value for AI in the next few years? Right. So, I think there is no big surprises for it's being used now. Right? I think it was really sort of reporting disclosure, carbon accounting. Right. These seem obvious use cases right now. Right. Because AI is greater at analyzing data and collecting data and drafting reports.

00;42;15;02 - 00;42;32;14
Andrew Jones
Right. And I think AI can really be a huge help if on the reporting and disclosure side with, you know, human oversight, of course, and human review and not just automating the whole thing. Right. I think it is useful. When we looked at the data from where our sustainably had saw the biggest opportunities in the years to come, it was clearly not the biggest prize.

00;42;32;14 - 00;42;58;06
Andrew Jones
And they cited much more ambitious use cases. I think where it's around now, AI optimizing energy efficiency in the buildings, right. And improving manufacturing, improving visibility into supply chains, enhancing climate risk analysis, which is very sophisticated. I think AI is a real big out there. And even yeah, some quite interesting cases that we're starting to see take shape in real life sort of around here using AI in product design.

00;42;58;07 - 00;43;21;09
Andrew Jones
Right. So, identify or even sort of help create these lower impact materials and reduce packaging and improve circularity and so on. So, I think there's a lot of excitement in the sustainability field that the ways in which I could be used in the years to come to, you know, generate really meaningful where it's like meaningful reductions in energy and waste and pollution.

00;43;21;13 - 00;43;35;23
Andrew Jones
I think there's still probably some work to be done in, actually, you know, how do you really advance that work forward? How do you upskill the sustainability team to use the technology effectively? Right. I think that's still a work in progress in the same way it is in all kinds of functions. Right. But I think it's coming.

00;43;35;26 - 00;43;54;25
Andrew Jones
Yeah, I sustainability value in the long run is probably going to come less from, you know, writing better reports and more from running better operations. Right. And I think, it's exciting as an observer, I hope we can get there. And I hope we can leverage these technologies for positive sustainability impact.

00;43;54;27 - 00;44;31;12
Marc Siegel
Yeah. Me too. Obviously, when you hear about things like potentially new materials and new products and improved efficiencies in terms of machinery and all that kind of stuff, and how do you design buildings, etc., that all sounds fantastic and very inspiring. So hopefully, this will continue this trend. And I guess I wanted to bring it back to the students that I see at Fordham all the time, and sort of the young accounting students that I talked to regularly who are a little worried about efficiency, meaning difficulty in jobs and less need for jobs.

00;44;31;12 - 00;44;50;28
Marc Siegel
And so, for students listening who are interested in sustainability or governance or responsible business, what skills do you think they should be building now? And I think about data literacy or regulatory fluency, communication ethics, systems training is as this is evolving and no need to hold you to this because it's going to change consistently over the next couple of years.

00;44;50;28 - 00;44;54;19
Marc Siegel
But what are you thinking about? What are you hearing and what would you recommend?

00;44;54;26 - 00;45;15;20
Andrew Jones
Yeah, it's a great question. And do you reflect on this a lot? You know, exciting I think to your point, this is, you know, this is an interesting time in the labor market as a whole and obviously an interesting time to be entering the labor market. Right. And a lot of similar assumptions are under challenge. Now. There's a lot of uncertainty about where certain professions will or certain functions or certain disciplines will evolve.

00;45;15;20 - 00;45;30;24
Andrew Jones
Right. But I think I think there are some core skills. I think that definitely students can and should be building. And yeah, I think you can't get away from the data literacy and analytical fluency. Right? I had given that these AI tools are here to stay, I don't think we all need to become data scientists.

00;45;30;24 - 00;45;55;20
Andrew Jones
Right? But I think we do need to build some general familiarity with data, what it can prove and what it can't prove. Right. And when the claims are running ahead of the evidence, I think any grounding there you can you can build for yourself will come in very useful wherever you take your career. I think more broadly, in particular, if you're interested in, you know, corporate sustainability, corporate governance, responsible business, I think that business, regulatory fluency I think is really helpful.

00;45;55;20 - 00;46;14;24
Andrew Jones
Right? You know, it's, you know, as we sort of been speaking through this discussion. Right, Mark, you know, there's a lot of disclosure rules, a lot of regulation, a lot of policy assumptions that are all shaping this field and will continue to shape this field. Right. And it's hard to stay up to date, but I think any sort of fluency you can build there, any kind of I guess broad view will becoming very helpful.

00;46;15;00 - 00;46;41;18
Andrew Jones
Because no doubt a few years from now, there'll be a whole new raft of policy stuff that we didn't see coming. Right. My final recommendation, I think this is particularly useful for sustainability minded folks is just being able to perhaps be a translator. What I mean by that is, you know, being able to move between like the technical expertise, which is really important and remain really important, but also being able to talk, you know, in the language of external stakeholders, of an internal stakeholders and investors and employees.

00;46;41;18 - 00;47;04;21
Andrew Jones
Right, and law and regulation, I think I think being able to explain complex issues enough to different audiences is really, really valuable. And I think some business acumen helps there. Also, those are probably quite broad recommendations. Right. But hopefully they are quite practical. And I definitely say don't chase labels like ESG right or responsible AI

00;47;04;21 - 00;47;19;17
Andrew Jones
You know, they'll work, will come and they'll go, but build durable capabilities, right. Whether it's in communications and judgment and systems thinking and data I think I think they'll serve you well wherever you take your, your career.

00;47;19;20 - 00;47;46;09
Marc Siegel
Oh man I couldn't agree more. Couldn't agree more. I think the points that you break, especially around some of those soft skills that probably don't get enough attention in the classroom, and that ability to translate between something technical to something that's much more accessible by a wider audience or a different audience. I think that that's huge. And whatever the case might be for your career, there's always going to be things that are deeply, deeply technical that need to be explained to the non-technical.

00;47;46;11 - 00;48;14;06
Marc Siegel
And I couldn't agree more that having that skill set of translating and communicating and being able to deal with other humans and explain to them what's going on is going to be critical, right? Because like you said, in the next three years, something else that's completely unforeseen right now is going to come up. And that's the other thing that I tell students and that you implied, might as well make it explicit, is that don't think about your career over a 30-year landscape and try to decide now what you want to do for the next 30 years.

00;48;14;06 - 00;48;22;12
Marc Siegel
Right. Just try to think about what do I want to do for the next three years and enhance my skills in these different areas. That'll just lead to more opportunities as the world keeps evolving quickly.

00;48;22;14 - 00;48;38;12
Andrew Jones
I love that that last point, Marc. You know, I think that's always been valid and now perhaps is more valid than ever. Yeah, it's sort of you can have these big visions for where you want to end up. And I think that's really great. Right? You need to have like a goal and be able to visualize it. But, you know, I guess where we started this conversation.

00;48;38;12 - 00;48;53;17
Andrew Jones
Right. So, our career paths are often non-linear and opportunities come and you take them right. And you don't necessarily know where they're going to lead. I think you have to be nimble, have to be agile, and therefore you may be focusing on what you really want to accomplish in the next few years is a really helpful foundation to do that.

00;48;53;17 - 00;49;12;11
Marc Siegel
I think that's a perfect call back to how we started the conversation. Look, Andrew, this has been really great. Thank you so much for joining me and for sharing all the great thinking that you've been doing in the realms of sustainability and AI and the intersections between the two. That's just truly a fascinating discussion. And thanks to everyone listening to People Behind the Impact.

00;49;12;14 - 00;49;26;01
Marc Siegel
If you enjoyed this conversation, please take a moment to subscribe to the series, write a quick review and share it with others who are interested in responsible business, sustainability, reporting and leadership. Thanks again for listening and we'll see you next time.