AI and Climate Change: What’s Next

Interview with Juvarya Veltkamp: Can Technology Save The Planet

Juvarya Veltkamp is the Green Economy Initiatives Manager for the City of Vancouver, working in this role for more than 10 years. She is part of the city’s dynamic task force responsible for the rapid adoption of a green economy and double-digit growth in green jobs as Vancouver continues to implement its Greenest City Action Plan.

We met earlier in March at “Women in Debate: AI and Climate Change – Can Technology Save The Planet.” The event was organized by by Narges Nirumvala and Women in Communications & Technology in association with International Women’s Day. The topic was provocative and tough. I was very curious how the debaters would pull it off, and these women blew everyone away. The organizer used a very effective debate technique in which one team argued for technology saving the planet, and the other argued against it. Super-knowledgeable, eloquent, passionate Vancouver women: Juvarya Veltkamp, Kiera Schuller, Lori Guetre and Sharon Sandhu provided the audience with education, entertainment, and enlightenment.

I couldn’t resist bringing some of this conversation beyond the debate room and Juvarya Veltkamp kindly accepted my invitation for this post-debate interview.

ZB: Do you have a goal, Juvarya? Life purpose?

JV: I’m from the UK. I grew up there, but also spent some time in Saudi Arabia, and because my family is from Pakistan, I spent some time there too. What I saw is that resources can be distributed in such different ways. In London, it’s a very affluent society. And then in Saudi Arabia, I saw how the Bedouins live in the desert, very in tune with their environment, and it’s a harsh environment. I saw what people need in different places and realized that we would have enough for everyone if we were more efficient with our resources and by being more effective with how we distribute them. So, my purpose became sustainability which is about not wasting and ensuring that we have enough for everyone, now and in the future. I like Kate Raworth’s model of Doughnut Economics based on a decent standard of living for everyone within planetary limits.

ZB: Just because it’s on everyone’s mind and you work for the city, before we move to our main topic, would you please share what you think about the Coronavirus and how are you coping with it? 

JV: On the city governance level, the situation has almost immediately moved from a health emergency to an economic emergency. It’s a good example of how everything is intricately linked. Of course, the first response is about health services and containing the pandemic. But, what this also brought home for me is the idea that people’s livelihoods are immediately impacted.

The medical response is one thing, and in Canada, it feels like it’s been based on a lot of evidence, but an economic response feels a bit more like we’re making that up as we go. Do we have the right models to think through the economic impacts? My husband works in the film industry, and as people stop coming to Vancouver to shoot, and most of the film industry is really globalized, that would impact our family and I know it’s already impacting other people’s families and livelihoods in similar ways.

ZB: Is there anything AI could have done to help us with the COVID-19 crisis? Have you thought about this?

JV: I am not in the medical field, but they must be using AI and predictive analytics to determine the spread and the medical response. Are we using it in the economic side? I don’t know, but the billions of dollars that Canadian Government announced is probably based on some algorithm. I don’t know where these numbers come from. It’s a very good point.

ZB: And now to our question: can technology save the world; can AI prevent climate change. That was a great debate. It was amazing to see how people in the audience changed their point of view at the end of the debate.

JV: I agree. I didn’t expect such a sea change.

ZB: When you were preparing for the debate, you didn’t know if you would be defending the idea that AI will prevent climate change or argue against it. So, what is your best pro argument right now and what is your best argument against?

JV: Climate is such a complex issue. And it’s not a crisis on its own. It’s interlinked with the climate system, the economic system, biodiversity, and the inequality. A lot to wrap our heads around as humans designing solutions, so we need a lot of help. AI opens up the ability to analyze and make predictions from so much of the data we have available and could never analyze effectively. I think it’s a critical tool at a critical time. Another way to phrase it: we probably couldn’t solve climate change without AI. Not as quickly as we need to.

For example, energy systems based on fossil fuels are the biggest contributor to climate change. Lots of changes are happening in the energy grid. By 2050, half the global grid is expected to be renewable. And to have a renewable energy grid, we need technology. If we have renewable energy sources, they’re very different from a traditional energy grid where you have one big client sending energy around the whole province. We’re talking about distributed energy resources, solar panels on everyone’s roof, wind turbines, cars charged on the grid. All these connected devices need to talk to each other. So how do we put all these different sources of supply together to make a virtual power plant, and then connect it to everything? You probably know Enbala is creating virtual power plants. It’s all about sensors, data and optimizing the energy flows through the day. We can’t have this new energy grid without digital technology. That’s one thing that AI facilitates.

Ownership of the grid is also a good question. Who benefits from the grid and from each individual transaction, who owns the data about the grid, and all the insights and predictions embedded in that data? I think it’s important. We could just say, well, one person is going to own all of this or we can use block-chain technology and distributed ledgers and take back control and create a democratic energy system. The AI has so many applications, not just in generating climate models, but also in integrating technology, smart mobility, and public transit systems. We need help with all of that.

ZB: Thank you, Juvarya. And what would be your best argument against it?

JV: AI is still a tool, so who wields the tool will impact the outcome. It starts with human intention. If you don’t recognize that, if you’re not asking the right questions, it won’t be able to give you the right answers. Do you want to save the planet? Or do you want to save humankind? Or do you want to save all animal life? Those questions have such different values embedded in them. Often, we talk about saving humanity, but do we want to save animal life and biodiversity? We’ve lost over half of the wildlife on the planet in the last forty years. And it makes it very clear to me that we need a solution for all life on the planet. So, the argument against is that technology is a great tool, but we must remember what’s the human intention behind it and how we’re applying that tool.

ZB: Thank you. What the city of Vancouver is doing to combat climate change? The biggest programs that you’re working on?

JV: In 2010 Vancouver created the strongest citywide climate action plan that I’ve seen. We’ve been focused on several outcomes and reducing emissions is one of the priorities. In the next 12 years, we’re going to have all new buildings in Vancouver built to a zero-emission standard. That’s one of the most progressive green building codes anywhere in the world. We’re trying to work with the industry to lower the carbon content of the building materials and we’re also looking to implement a retrofit policy to improve the efficiency of some of our larger existing buildings.

This drives market transformation. We have built a brand that is very valuable around the world. Our expertise in planning, design, buildings, mechanical engineering, building envelopes, etc. is in demand all over the world. In Vancouver, the industry and policymakers work together on setting progressive targets and creating solutions. Some people say, businesses just don’t want regulation. It makes everything expensive. But I think businesses are just looking for predictability. They want a level playing field. And when they know the rules, they can see how to excel within those rules. The leaky condo crisis that happened in BC is an interesting story. By fixing the leaky condos we’ve developed expertise at building envelopes and in engineering very energy-efficient buildings around the same time the Kyoto Protocol was announced. Lucky for us, we already had this new expertise at hand looking for the next big challenge.

The city of Vancouver has a goal to be 100% powered by renewable energy by 2050. Built environments and buildings are large part of it, but we’re also tackling transportation and fossil fuel vehicles, our next big source of emissions. We’ve done a lot to put electric vehicle charging infrastructure on public property, but also to facilitate it on private property. This includes requiring developers building parkades to dedicate a percentage of new spots for EV charging. At the provincial level, we’re saying that by 2024, you will not be able to buy a new fossil fuel vehicle in BC. I think that’s a strong signal to the automobile sector.

Part of our green economy strategy is also our support of cleantech entrepreneurs. Big institutions, like the city, often find it very hard to work with innovators and all the latest technology that’s coming up. So, we’ve created the Technology Deployment Network, where we help large buyers like the city of Vancouver or New West, or Oxford Properties so that through our program they can work with local innovators and learn about and test new technologies.

ZB: Thank you, Juvarya. Now that we discussed the opportunities, can we talk about the risks?

JV: AI is a game-changer. It has the ability for a big change and it’s already embedded in so much of what we do. It can bring great transformation if deployed at scale. It can also have massive unintended consequences. So, designing AI-based systems, we need to be very aware and be able to course-correct rather than build on shaky foundations.

We’ve already discussed the intention behind AI. Similar to this is a bias. For example, Google got in trouble a while ago when they were showing highly paying job ads much more frequently to men, than women. That shows you how bias can get embedded in an algorithm. So that’s one of the issues, how to design the algorithms so that they’re not perpetuating bias in the system. Another issue is surveillance capitalism. You know how we used to think of economic production. We work, work creates value, we get paid for that work, and that’s how the capitalist system is based. But in “surveillance capitalism” value is being made not of our work, but of our daily routine. I go to work, and Google knows I went to work, or that I purchased diapers or something else. Our regular daily behaviors are generating data that creates value, that companies then sell for the predictive value of that data. You can get into data privacy issues, but assuming everybody doesn’t mind that data about them is being collected, who gets the value? Right now, it’s the company that has the data and sells it. We don’t benefit. Is that the right way? Should some of that value come back to society or the individuals? We have no rules about that. If Google wants to donate its Google Earth or Google for Good or whatever, that’s their choice. But we, as a society were not given that choice.

Same with climate change. We didn’t set carbon pricing, didn’t value the ecosystem, and it created imbalance. If this was thought through, we would have made the system roll out in a different way. If we had to pay for carbon pollution from the very beginning, people would have been much more efficient and not polluting as much. Same with AI. If you must contribute some of this value to society, then things will roll out in a different way. But right now, we just don’t have rules.

ZB: How can we, as a society, protect ourselves? Where do we start?

JV: I think a great place to start is a “Do no harm” pledge within the AI community. Also, let’s get rid of the bias. We can do that. If you look at sustainability, it’s gone much further than “Do no harm.” We now talk about designing business strategy and purpose that is making a better world. Just like you do at Freightera. It’s not about what’s the minimum we can do. It’s about how do we actively benefit people. And I think that should be the next step on the AI side. Even if your AI isn’t specifically addressing a climate or social outcome. How do you make sure in all your operations that you’re still benefiting the society? I think that’s the next step.

From my experience working with the building sector, we need a multi-stakeholder approach. Some of the most effective tools we had in green buildings, for instance, was LEED certification developed by the designers, the engineers, and policymakers. We designed it together to be able to say that when we put this mark on a building, all of us agree this is a green building. It was also a way to communicate to customers that they are buying a certified product. It was developed hand in hand with different parties working together. I think we’ll need a similar approach to agree on frameworks and tools for developing ethical AI policies.

ZB: I noticed that you mentioned capitalism a few times. I think it’s very clear that our ideologies and our values play a huge part in the development of AI. Would you please comment on that?

JV: For me, the biggest example would be, Uighurs in China, where the government is now using facial recognition and other AI technologies to track and control a whole minority population. You can see, when you apply a different value, the stability of the whole nation becomes more important than individual freedom or even the freedom of an entire Muslim minority. So where does an ethical framework come in? How should we be applying it in different political systems? China is not a pariah state. It’s their political system. They’re part of our global economic framework. If this is what their government decides to use technology for, who are we to say what they can do? I don’t have an answer to that. But it’s a very important dilemma.

Reflecting on sustainability, there are companies that operate in Bangladesh, for instance. There were a lot of factories with terrible human rights practices, terrible safety practices that led to deaths when one of the factories caught fire. And a very real question for textile companies is, what do we do? Do we just pull out of Bangladesh? Or do we stay and engage? And if we stay and engage, what are the parameters and how do we determine our ethical framework within the communities that we operate?

ZB: You mentioned during the debate the use of natural language processing to identify sustainable investments. Would you please tell me more?

JV: You know how companies are easily assessed on financial metrics. We’ve developed a whole accounting system for hundreds of years, we’ve refined it, so we have the metrics indicating financial performance. In sustainability, we’ve also got metrics to identify sustainability performance. Interestingly, more and more, sustainability metrics become proxies for financial stability. Companies with great sustainability practices actually perform better in the long run. There is a whole world of sustainability metrics known as ESG; environmental, social, and governance metrics. You can measure things like carbon footprint or emissions per dollar revenue. You can also measure diversity; how many women are on the board. There is so much ESG data, and navigating that data is a whole field on its own. Some companies are starting to use technology to navigate that data, organize and analyze it, so we can better understand the impacts of ESG on financial performance. We can then invest in companies based on that performance.

Some companies don’t disclose ESG metrics. It’s not regulated. They don’t all report in the same way. And that’s the problem. AI is helping to solve it with natural language processing. You can go on the web, you can look for articles, news reports, reviews, comments, anything about the company. True Value Labs is working on analyzing this data and looking at how it can be used in creating an investment portfolio mix based on these metrics. It’s like holding a mirror up to these companies to see how they perform and how it aligns with our values or not.

ZB: Thank you, Juvarya. Could you tell me how this affects jobs?

JV: I look at this in terms of fossil fuels. We want to make 50% of the grid renewable in the next 30 years. That’s a big transformation. A big change in the kinds of jobs that will be available. So, in my work, we talk about a just transition for fossil fuel workers. The change in the industry is not the workers fault. They’ve invested in and it’s where their pensions will come from. How do we ensure that we transition without losing jobs? As we’re seeing now with the Coronavirus, when people lose their jobs, it impacts the whole economy. It’s a big deal.

We need to plan for the transition so that we don’t end up with the massive social disruption. You can retrain people. You can early retire people. You can relocate people that want to be relocated. A lot of oil and gas workers are used to moving around following the jobs. The renewable energy jobs might not always be in the same places where the oil reserves were, so we need to think about the HR strategy for them.

We are talking about different types of people, different types of skills, a different type of social contract. If you look at Uber and Lift the gig economy companies, I heard the other day that 10% of people in Canada are working in the gig economy. These people don’t have benefits. They don’t have long-term employment. They might not even be paying into EI. Their social net is very different.

There is also a scaremongering that AI will take away jobs. Well, only if we, as a society, let it happen.

There are so many good things that we could work on together.

Juvarya Veltkamp was interviewed by Freightera Co-Founder Zhenya Beck.

Click here to read more about Freightera’s green mission.