Responsible AI: A Conversation with Elisa Moscolin, EVP of Sustainability, Sage

Transcription

David Stearns:
Welcome to BSR Insights, a series of conversations on emerging and cross-cutting business, economic, and social issues. Drawing on BSR's expertise for more than three decades of leadership and sustainable business, we'll help practitioners and decision-makers to navigate today's increasingly complex world. I'm your host, David Stearns.

In our previous episodes, we chatted with Lindsey Andersen and Lale Tekişalp, members of BSR's Responsible Tech team, about new and emerging AI technologies being deployed by companies around the world and how to ensure that AI deployment is managed responsibly. We also chatted with BSR's Richard Wingfield and Will Manley from London-based law firm, Slaughter and May, about the geopolitical and regulatory environment companies are operating in, and the challenges they're facing due to diverging policies and shifting stakeholder expectations.

Today, we'll be diving deeper into the nexus between AI and sustainability, a topic which BSR covered in a report we released last year titled, “Harnessing AI in Sustainability: Emerging Use Cases.” You can find that report on the BSR website, www.bsr.org. So here to help us do that, I'm joined by my colleague, Laura Gitman, Chief Impact Officer of BSR. Welcome, Laura.

Laura Gitman:
Thanks. It's a pleasure to be here.

David Stearns:
And Elisa Moscolin, Executive Vice President for Sustainability and Foundation at BSR Member, Sage. Welcome, Elisa.

Elisa Moscolin:
Thank you. Hi.

David Stearns:
Hi there. So I thought Elisa, maybe you could start. Tell us a little bit about your role, maybe a little bit about what Sage does, and your role in sustainability at Sage.

Elisa Moscolin:
Sure. I'm responsible for sustainability and the foundation at Sage. We are a software, accounting, and HR services solution for small-medium businesses. We are one of the biggest tech companies headquartered in the UK—very proud UK company—but we're global. We actually have a lot of operations in the U.S. My role is about really making Sage more sustainable, responsible, and most importantly, bringing sustainability to small-medium businesses and helping them through their own journey.

David Stearns:
Very much looking forward to hearing more about how Sage has been working with AI and how this has been part of your company's sustainability journey. But I'm going to start with Laura. I mentioned last year's report on harnessing AI and sustainability, and it's been almost a year since you and the team chatted with CSOs around the world about how sustainability teams are using the technology, from leveraging it to boost day-to-day efficiency to reporting and disclosure and strategy and risk mapping. 

Obviously, AI is advancing and developing at a very rapid pace. More and more companies are adopting it. Would be curious to hear your perspective on how CSOs might answer the question today around how companies are adopting it. But first, for those who haven't read the report, what were the key takeaways from your engagement with CSOs on the topic? What were the main things that you heard from those conversations?

Laura Gitman:
It was an interesting exercise and we interviewed 20 CSOs last summer and published the report in September. What we really wanted to understand is, how are CSOs and core sustainability teams using AI in their day-to-day work? And so, we were not really thinking about how it might be used for broader purposes in terms of the company's overall deployment of AI, and we weren't even looking at what maybe facilities teams, or energy management or procurement teams, were doing that might have sustainability benefits, but really, how does it impact the day-to-day work of a sustainability team? 

What we found is that it was extremely nascent and really uneven, but we identified at least four categories of use cases where I would say companies were starting to dabble and think about the potential use cases and upsides. Those were first and certainly the largest category, which was just day-to-day productivity and improved communication—so really the same way AI might be used by any team, not sustainability-specific. Of course, sustainability teams are small, often resource-constrained. That can be a huge asset for some of those small teams. And so, productivity and improved internal communications was a key use case. 

The second was around reporting and data compliance. This is a big part of the function of any typical sustainability team. It's often quite time-consuming and can be burdensome. And so, a lot of teams were starting to explore how to use it to improve the reporting process, all the way through actually helping with the drafting and editing process. Many were also using it to do gap analysis from their previous reports against various new or emerging external standards, or external regulations, to help inform reporting for the year to come.

The last two categories were much less common. One was strategy and risk management—so starting to really have AI be a thought partner to help identify key priorities, anticipate risks, maybe about investing in a new market or a new product, and what they might be able to anticipate. And then last but not least, stakeholder engagement—so thinking about how to use AI to kind of anticipate what different kinds of stakeholder profiles might say. Pretend you are a major NGO. Pretend you're a media organization. Pretend you are a regulator from this agency and really starting to get good feedback in a way they really never would've been able to get before. You can't always have those direct conversations, and that has been a really valuable form of using AI. 

I think the other thing that we certainly saw was that there are a lot of opportunities that are going to have sustainability impacts, but that are not driven by the sustainability team. And there is a lot of excitement on that around energy management, around logistics, and around R&D and design for products especially.

David Stearns:
That's really interesting. Thank you, Laura. I think it's been around nine months since we published that report. Obviously, things are moving really quickly. You've been continuing to have conversations with our members. How have things changed in that intervening period? Have you noticed any shift or more advancement of the use of AI? You mentioned companies were rather nascent when you first spoke with them nine months ago. Have you noticed any increased capacity or competence in utilizing this technology over the last intervening nine months?

Laura Gitman:
Yeah. I mean, we certainly haven't gone back to those exact same companies, but I would say candidly, it's still pretty nascent and for most companies, maybe it's not the full nascent stage, but it's still very much a pilot phase. So there's a bit more exploration, there's maybe more comfort, more experimentation, but we're not seeing companies that sort of already have a very clear strategy and implementation plan around all of the ways that they are internalizing and using AI. It's still in that, “let's test and experiment” phase. And I can tell you, we did a poll of 40 of our members just in April and the majority of them were still only using it for productivity and efficiency. They hadn't gotten to the other three use cases yet. Where we have seen some changes, and one of the areas that we really highlighted companies should start with, is the preparation phase.

There were two things that we said really need to happen before you can start to see the benefits and have the use cases. One is data modernization. It's really hard to use AI if you don't have good inputs to put into the AI about your business. You can't really use it for reporting or analysis if you don't have a clean data set. And so, a lot of investment in that first phase around data modernization, and I'd be curious if Elisa has experienced that. 

And secondly, setting up the foundation of how to use it responsibly. We spend a lot of time helping companies think about that, and I know Lindsay and Lale probably talked about that. What are the policies? What is the governance? What are the risks that are going to come up for your company depending on where you sit in the value chain? So I think we're still in the phase where companies have to invest in that infrastructure before they can really invest in some of the more exciting use cases that we've identified.

David Stearns:
Thanks for that, Laura. I am curious to hear how any of this might have resonated with you, Elisa. Maybe you can start by telling us a little bit about how Sage as a company has been adopting AI, and how that's influenced your approach to adoption for the sustainability team.

Elisa Moscolin:
Sure. And if it's okay, I'm going to build on what Laura shared because I think it's very interesting and I definitely see a trend of CSOs struggling to engage with this technology. I just want to put that into context because I've been talking to CSOs around the world, of big and smaller companies, and I think the challenge that this community is facing is that we are navigating and trying to, frankly, survive a very volatile geopolitical environment. We're under extreme pressure and many of us in certain geographies are for sure feeling under attack and having to defend the base.

I would also say that there is an interesting paradox, where a lot of the attention has been grabbed by AI and to some extent, getting funding and attention and the airtime is harder than it's ever been. So that, I think, has pushed many CSOs and many sustainability teams into the trenches and on the defensive, which I think is a problem. My argument has been that we unfortunately cannot afford that. When it comes to AI, we can only be at the forefront. And we're going to maybe touch upon this a bit later, but I think that adoption of AI for sustainability people is not enough. We need to really master it. And there are several reasons as to why I think that, but we can come back to this point. I don't want to elude your question about our journey at Sage and what are we doing.

I would maybe start by giving a bit of context because AI at Sage is a bit different. So as I mentioned, we provide accounting and finance software solutions, and HR software solutions to small-medium businesses. And in our world, for our category, an almost right answer is a wrong answer. When you do accounting, when you do people's tax returns, when you need to close and reconcile a balance sheet, almost right is not right enough. So it's a very different type of AI. We need AI that generates certain things. An answer that sounds about right is not an answer that we can trust. So, we need control—our customers need control, confidence, and accountability in the answers they get and how they deploy AI, which is very interesting and exciting because I think for my team and I, from a sustainability perspective, that puts a focus on trust.

And of course, trust for our category means the various things I've mentioned—so control, accountability, transparency, explainability—but if you take it a bit further, it is also about responsible AI. It's about that transparency, that confidence that we will deploy it in the right way, that we will make sure that the outcomes that it delivers ultimately serves our customers and our business, and if you want to stretch it more broadly, the economy and the society. So, I find it a very fascinating and interesting place to be. I think we live in a world where the race for AI is becoming a geopolitical imperative and for companies, more importantly, it's a competitive imperative. We cannot shy away. It's interesting—I've got climate scientists on my team and we often have to battle with some of the trade-offs and conundrums of how do you deploy AI responsibly, how fast do you want to innovate, and what are the emissions, for instance. But you need always to balance it with the innovation and the competitiveness imperative.

I would say it's a very interesting moment to be alive, and I think I do relate to some of the challenges of my peers around prioritizing AI mastery over the hundred million things that we still need to deliver. But I also believe that if we don't, we'll become very quickly obsolete. So I think it is an imperative for sustainability teams as much as it is for the rest of the world.

David Stearns:
It sounds really like trying to thread the needle, right? Balancing the AI strategy, the need for the innovation, with the sustainability strategy in light of all of the pressures that you've cited—the geopolitical pressure, the pushback, the difficulties that companies are facing in that realm. You talked a bit about the importance of making sure that both of these strategies can come together or converge in a way and proceed. Can you talk a little bit about how you have managed to thread that needle? When you have the climate scientists in the room, you have your innovators in the room, how have you managed to bring them together and align them in some ways?

Elisa Moscolin:
So look, I think that if we take it up a notch, and then we can deep-dive on climate and other aspects, I feel the deployment of AI and AI strategy is the first of trade-offs—experimentation versus scale, speed versus governance, automation versus human touch, innovations versus ethical responsibility. The list just goes on and on. What I’ve found most fascinating and encouraging as well, is that my team are masters of trade-offs because if you think about it, there are a lot of parallels and transversal, transferable skills in sustainability that you can apply to responsible AI. We're actually using a lot of what we learned in our sustainability journey to navigate the AI era. And I can give some examples which also then explain how we're navigating these trade-offs, right? You start from the very beginning. Responsible AI means so many things for many people. And so does sustainability. You ask what is sustainability and everyone will give their own interpretation, from recycling to investing in social causes. So, you have a very mixed bag, right? Same with responsible AI. 

What we're learning as a company, and also through the help of my team and across functional collaboration, is that it goes back to materiality. So very similarly, as you don't define your sustainability strategy out of thin air, when it comes to responsible AI and the trade-offs and how you navigate, well, you look at what is most material and most important for your business. What are the impacts that AI is going to have on your business? So, what are the biggest risks, opportunities, and impacts—positive and negative—that it's going to have? And on that basis, then you can look at what the principles you need are going to be to really prioritize, and the interventions that you need to maybe wrap around stronger governance to make sure that you manage them a bit more closely.

One of the challenges that we've been coming up against is again, very similar to sustainability, where once upon a time, we used to talk about responsible business. So you can hear the alliteration there. So we are often seen as the people who put the brakes on things, who tell an innovative private team to stop, to go slower, to think about longer term. In reality, as we've proven a bit also with the sustainability agenda, that's not the case. And the same goes with responsible AI. I see responsible AI, as well as sustainability, as the fast track—the speed lane—where if you know where you're going, if you have a well-built road, you've got the confidence the tarmac is going to hold and that there are guardrails so that you know where you're going, and you know where it's taking you, then you can go fast. And you've got brakes that you can at any point pull and slow down if you need to.

That's where I see responsible AI as, really, the fast track—what gives us the confidence that we can innovate, especially in our category. Getting it wrong has a high cost for our customer and, ultimately, our business is a business of trust. And that's true for many other businesses. Think about people who deploy AI in pharmaceuticals. I mean, I was recently in San Francisco because we had our Sage Future event where we meet our customers, and I was in a self-driving car and thinking, "Oh my goodness, I hope this company has very good responsible guardrails."

So I think that there are many applications. In reality, we’re not the only high-stake company that really needs to get it right. But that's how we're navigating these trade-offs. We’re looking at what's truly material and how do we make sure that these guardrails we're putting in place, these controls, this deep understanding of the risks, impacts, implications, and opportunity, help us guide and take the best possible decision that leads to the best outcomes for our customers. And ultimately, it boils down to a license to operate and our credibility in the market. So I think it all makes perfect business sense.

Laura Gitman:
I love that metaphor around the road and driving. I have to think a bit more about that, but it's almost as if the responsible AI foundations are, as you say, sort of the guardrails. I think the other thing you're hinting at is that in some ways, the sustainability team can also be a bit of the navigation system, so making sure that the direction in which it's all going is able to think long term. Sometimes, the innovation sort of happens in the near term and doesn't necessarily have the liberty to think about the long-term societal implications. The sustainability team has that experience, but also has the knowhow to help a company think through those trade-offs, but also to think about, are you solving real problems for the future, or are you creating problems for the future? And how do we think about that in a way that balances that and as you say, then enables it to go fast.

Elisa Moscolin:
A hundred percent. I think you're spot on, Laura. And you and I have spoken at length about the crucial role that we believe sustainability leaders have in shaping and deploying AI responsibly. I feel we have a once-in-a-generation opportunity to get this right. I was saying that it's almost like sitting and watching a movie I've already seen. It all looks very familiar. When I look at how we're going about deploying AI, especially in certain applications, it feels a bit like when many, many years ago, companies would go on, do their business, sell things, pollute a river. No one would really care because that pollution, that river, wasn't attributed to them. But eventually, we know someone's going to turn around and go like, "Hey, that river you've polluted, that externality, you've got to do something about it. That’s not allowed anymore, so go rewind everything and put in controls," and then it's going to cost a lot of money.

If you think about all the retrofitting we need to do to transition to a green economy, that's the best example I've got, wouldn't it have been smarter to get it right from the get-go. To some extent, it feels like we've got this opportunity now. And don't get me wrong, I am deeply aware of the trade-offs that we need to navigate. We live in the real world where, as I said before, adopting and excelling in AI is a commercial imperative, and if you lose that battle, you're out of business. That's why I think it is crucial that we make responsible AI an engine for innovation that will bring results in the short term, but also in the long term. The two things I believe can go hand in hand if done smartly and calibrated. It's similar to sustainability. So it's really a movie that I've already seen, and I hope we can borrow a lot of the lessons here and get it right at the first get-go.

David Stearns:
We've heard from a number of companies, in talking with them recently, that some are setting up responsible AI task forces and finding out that the sustainability team does not have a seat at that table. So that's been one of the first places we've been really pushing for members, to make sure that the team has a seat at that table. I assume, based on your comment, that that’s something you'd be strongly advocating for as well to make sure that the sustainability team is part of that. 
We've been talking at quite a high level. Are there any specific examples or anecdotes that you can point to where your approach to responsible AI strategy, or the governance of AI, has produced a particularly positive result or advancement? Is there a specific example that you can say, just to sort of bring it to a more tangible level for our listeners—something that you're particularly proud of or that you think you'd love to see more of within Sage?

Elisa Moscolin:
I'm going to link to the previous comment. So my team and I have a seat around the table when it comes to responsible AI. We're proud of it. We have cross-functional collaboration because again, similar to sustainability, it sits across functions—it's risk, it's legal, it's the products team, it's the cybersecurity team, and it's us. We all have a voice, and the lines—it's interesting—blur quite quickly because then you have to look at the whole AI life cycle. So it's truly a very big cross-functional collaboration effort. What I'm very excited about is that we've recently launched, when I was in San Francisco, Beyond the Black Box. This is a partnership, together with PwC. We are going to soon be trying to convene more friends around the table, because what we'd really like to do is try and define what trust means for our category and provide especially small-medium businesses that confidence that they need to adopt AI that they can trust.

I'm very excited that my team and I are at the heart of this and that we are working with other functions. I must confess, I'm on a very steep learning curve with everything that comes to responsible AI, but I'm loving it. And as I said before, I think I'm bringing a lot of lessons from my years of experience in the sustainability field. I think that ultimately, this is a great opportunity to really drive that confidence and put the controls in the hands of our customers. So what I hope we'll be able to see as we go on and develop this partnership is that we can empower small-medium businesses to feel in control—to understand, what are the things they need to look for and when they deploy AI, what is the checklist that they need to have ticked off to have confidence in their deployment and use of AI?

We've done some insights and what we found out was quite interesting. AI adoption is not a technology barrier, it's a confidence gap. If we can bridge that confidence, we can make this technology more accessible to everyone but also, we can make sure it's deployed in a responsible way. So, the two things really go hand in hand. Then personally, I'm very excited. My team and I are doing a lot of work, and as I said before, I live by the mantra that we cannot only adopt the sustainability—people must master it. And I think that's because, as Laura alluded to before, sustainability teams are historically very small compared to the size of their agenda. So I think this is a fantastic opportunity to really increase our impact and effectiveness.

Second, I think that AI truly changes the game. It can solve problems that we've been struggling with for years. In my space I've been struggling with, how do I bring sustainability to small-medium businesses? It's a hard crowd because when you are a small business, you're focused on the day-to-day. In the morning, you are the Chief Risk Officer. At lunch, you are the Chief Finance Officer. Around 3:00 PM, you are HR, and in the evening, you're the delivery person. So it's impossible in that context to also ask them to become the sustainability expert. But I think AI is changing that dynamic and is opening up opportunities that we didn't have before. So, I'm really excited about technology. And last but not least, and again, Laura, you alluded to this, right? I think sustainability leaders have this crucial role to play in deploying AI responsibly.

I hope it doesn't come across as egocentric, but I do think that the sustainability community has an important role to play. And we've got lots of lessons that we can borrow from the journey that the sustainability discipline has gone through. Again, I'm on a very steep learning curve. I've done different courses and I'm learning on the field, day in and day out. I think this is essential, but I think everybody is. The technology is evolving so quickly that it means this steep learning curve will continue to be steep for a while.

David Stearns:
Elisa, have you heard feedback from some of the small- and medium-sized companies that you've been working with? What kind of feedback are they giving you? Is it helping you to then make further enhancements so that the product becomes even more useful to them in a responsible way? But I'm just curious to hear what kind of responses you've received.

Elisa Moscolin:
Yeah. We've got some insight and I'm happy to share in terms of for our category, how people are thinking about AI. I think what was most surprising to me for sure was this confidence gap that we believe we often talk about, "Oh, AI was going to save us so much time." And then we find out that our customers, specifically, erode those time savings by having to check that the answers are correct. And I mean, thankfully, our customers do that, but how many people actually do? I mean, I don't know about you, David and Laura. I often do take it for granted. I'm like, "Yeah, that sounds about right. That's convincing enough." But for us, convincing enough is not good enough. I think what our customers are telling us is that they want to adopt it, but they need the confidence. They need to be able to trust it.

Ultimately, if you think about a CFO of a small-medium business, or someone with a small business using our products and our AI to file their tax return, they end up in jail if we get it wrong. They can get fined or their career, their credibility, is at stake. For me, it's been an eye-opening realization that actually, when I use AI, I can't blame AI if the answer I'm giving is rubbish. And so, that explainability, the traceability, that transparency in the outputs has been really interesting. And that's what the Beyond the Black Box that I've described is, it’s that confidence. But then from a sustainability perspective, again, I was recently in the US—tough crowd at the moment—but actually, what my reflection coming back from that trip has been, is that I think we need to move away from just positions in generic sustainability climate.

And this has nothing to do with AI, frankly. It just has to do with, we need to make it real. When we were there, we tested a number of AI-enabled sustainability solutions. On the ground, we had customers playing and looking at use cases, and what was interesting is that if you talk to customers about climate change, they can't really stick for too long with that conversation, or they go straight into, "Yeah, we do some recycling." But if you're in California and I started talking about how, for example, we're partnering through our foundation with some very innovative entrepreneurs that are lowering the wildfires, oh, that becomes interesting. If you're in Florida and you're talking about climate change, you're probably not going to make inroads, but if you started talking about floods and home insurances and insurability of certain properties, that's sustainability. It's just more precise and more kinds of factors.

So, I think truly, it's time that we make this real for our small-medium businesses. We talk about carbon footprinting and what we're starting to actually reflect a bit more on is, how do we turn that into savings? I mean, we live in a time when energy price volatility is the real problem for small-medium businesses. How can we help them navigate these disruptions? So, bringing abstract concepts to reality and using AI is the key that can unlock the potential of sustainability for small-medium businesses, solving their day-to-day pain points and issues. I think that's what makes me most excited about AI.

Laura Gitman:
Elisa, what I was going to ask earlier, and I love the excitement—I know you've been thinking and talking about this and you see this. I love when you describe it as this once-in-a-lifetime opportunity. What are other CSOs saying? Are you getting agreement? Are you getting skepticism? Confusion? What's been the reaction as you've tried to take that message to your peers?

Elisa Moscolin:
What I'm hearing is that everybody's on the fence at the moment. I do understand that and I'm not immune either. So, there are a few challenges. The comments I get typically is, "Look, I'm surviving. Budgets are being cut. My team has been under pressure. There are lots of regulatory reporting requirements, especially in Europe. So we're just keeping our heads above water." And I think that's a real challenge. However, my counterargument to that and also to my team has been, "Yeah, but AI is part of the solution.” So, unless we make capacity, unless we kind of try and spin two plates and get on board on this journey, we won't be able to get out of this. So it does require a discretionary effort right now, but I do think it's worth it. Then, I think the other challenges that we face, which are common to any other business areas, is not only capacity and capability and skills, it’s how do you avoid getting stuck in pilot purgatory—which is when we have a use case of proliferation, without actually scaling. How do we make sure that you actually start seeing the benefits?

And I think my journey, in the past six months especially, has been this very big penny drop. As I said, don't do AI for the sake of AI. And it seems obvious, but you cannot believe how much of this is going on. I've called it AI sobriety to my team because I think I've been guilty of a bit of a hype myself. So I came back from parental leave and discovered the world. I felt I was away for a hundred years, not one year, when I realized how much had changed. And now there was some tech that was writing my emails and I was like, "Whoa, this is amazing." So, I kind of jumped into the hype myself and said to my team, "We've got to embrace this now. Adopt AI." Then I came to sobriety myself. As I said, I've also invested a bit in training, in reflections, and in talking to other leaders, other companies. We came quickly to the realization that probably the biggest threat to AI is AI itself—is AI for the sake of AI.

And what I mean by that—and I'll give you a couple of real examples—looking at reporting, Laura, you mentioned that indeed, reporting is one of those areas that lends itself very well to AI. We started to explore, “Okay, so we can have an agent do this and have an agent do that.” And before we knew it, there were 20 agents and we had coded our inefficiencies. We had to stop and go back to the drawing board, starting from the very boring basics of a transformation mindset and going like, "Okay, what does good look like? What's that future state look like? Okay, what is today? What is tomorrow? What is the gap, and what does it take?" And it goes back to some very boring stuff, like what data do we need? When do we need it? Who has it? Where does it sit? Do we really need it? How often do we need it? So all those questions. And then, we learned quite quickly that some of the problems were not solved through AI, but were solved through very old-fashioned good governance and putting processes in place. And also, we realized that we were coding our inefficiencies.

I think there is a misconception that we can just have infinite agents. Agents are like an FTE. Very soon, we're going to have a price attached to them—that they don't come for free. So you better make sure that they are as efficient as they can be. The big realization was not only to stop doing AI for the sake of AI and treat it as a truly genuine transformation project, but also to truly do design thinking. What I mean by that is, if you design or try to do AI as a bolt-on, you're just coding the easy state. Whereas what you really want to do, and what we're doing is, you have to think fundamentally differently. In the reporting world, once again, we're doing all the boring work that I'm describing, which is the foundation if you want to do anything more shiny with AI.

But also, we're realizing that not all the solutions are necessarily what you want because everybody wants agents now, but it's not just agents. AI is many things. You have deterministic models. You have automation. You have so many others. And I think different problems will have different solutions. So, partnering with IT, partnering with people who actually understand a bit more than me on what is the technological solution that solves that certain problem, is important—and most importantly, thinking about what the future is going to be like. It's not going to look anything like what we have today. At the moment, for instance, what I'm advocating for my team is working on this ESG intelligence layer. So in the future, I see reporting just as a byproduct, as something that this ESG intelligence data layer produces alongside many other things. And I see that being dynamic. I see that being algorithm to algorithm talking, because what's the point today of producing a report really? It's not going to be read by a human being.

Laura Gitman:
It's going to be read by AI. Yeah.

Elisa Moscolin:
Exactly. That's exactly it, Laura. What I've been asking—what my team and I have been asking—is what's the future then of reporting? Because otherwise you end up creating a whole lot of work for fantastic AI solutions, an automated solution that does a report to only find out that actually, no one really wants a report anymore. That's why we came up with this ESG intelligence layer that can do many things. And I think this is true for many areas of sustainability—this deep kind of transformational thinking, redesign, really. Think about how it's going to shake the core foundations of what you do and only then, look at what are the solutions you can build. Otherwise, I think the biggest risk I see is going to do a lot of stuff that then we need to unpick and do all over again. That's going to be a big cost for companies and teams, and something we cannot afford. So lots of learnings. As I said, very steep learning curve.

David Stearns:
You've given us a lot to think about and a lot to chew over. I really appreciate that. Laura, any final thoughts you might want to add? Or maybe talk a little bit about some of the resources that BSR has developed that we can point our listeners to to help them as they continue this AI sustainability journey.

Laura Gitman:
Yeah, thank you. And thank you so much, Elisa—so much great insight that you've shared with us. I think I just want to reinforce your message around the opportunity side of AI and the opportunity side for sustainability objectives. That opportunity is not about making things a little bit more efficient—making our processes or communications more efficient—but really thinking differently and having tools that we didn't have at our disposal before to really solve some of the biggest systemic sustainability challenges. I think right now, we're still in the bread and butter sort of basics of AI, but I really hope we're going to get to some of these bigger solutions. They are dependent on us having, as you say, sort of the guardrails and the nav system. They're dependent on the right foundation, but I do think there's an incredible amount of opportunity when done well. So, I hope as a field, we can embrace that. 

I'll just say, BSR is really happy to partner with members wherever they are in the value chain. We have a number of members who are developers of AI, or who are deployers, and we've been working with them on a lot of the principles around responsible AI governance and implementation. We have a new group that is looking at the environmental impacts of AI but also, there are many companies for whom they don't even know how they're using AI. They don't really know what the big issues or risks are. As you say, Elisa—the need for a real materiality assessment to understand what matters to your particular business and what are the risks or opportunities going to be? And so, we're really happy to use our experience. We've got teams that have been working on these issues for 10 years, which feels crazy in the world of AI, but the teams that work with the developers started a long time ago.

So really pleased to bring those learnings, but then also to work with the companies that are new to this and literally just want to ask, “I don't even know how to write a prompt for reporting purposes.” We get those questions as well. And so, really pleased to partner wherever you are in that maturity scale, and thanks for the time.

David Stearns:
Thank you, Elisa. It's really encouraging to hear the thoughtfulness, and to borrow your word, the sobriety with which you are adopting or tackling these issues. I feel like if we came back and chatted with you in a year, the learnings would be exponential from where they are today. So maybe we can reconvene next year at this time and hear what new insights you've gained.

Elisa Moscolin:
Let's make it today, then.

David Stearns:
Okay.

Laura Gitman:
Sounds great.

David Stearns:
Thank you both, and thank you to our listeners for joining us once again for the BSR Insights podcast. I'm David Stearns. Thank you for listening.

Thanks for listening. For more in-depth insights and guidance from BSR, please check out our website at bsr.org, and be sure to follow us on LinkedIn.