Michal Nachmany: Unlocking Better Climate Legislation With AI

Written by the Patrick J. McGovern Foundation team.

When fighting climate change, knowledge is power. Climate stakeholders, from policymakers to business leaders to civil society, can collaborate more effectively and develop better policies and solutions when information is accessible and reliable. But what happens if the information available is a deluge of unstructured data from disparate sources?

Dr. Michal Nachmany saw this problem and decided to build a solution. As the founder and CEO of Climate Policy Radar, she leads a team that uses machine learning and generative AI to clean and organize climate legislation data across geographies and jurisdictions. It’s an approach that combines human intelligence with artificial intelligence. But Climate Policy Radar is more than just a one-stop shop for research and knowledge sharing. It’s rooted in a commitment to justice. Climate change disproportionately impacts marginalized communities and Climate Policy Radar ensures they have free and open access to the information they need to combat it.

Today, Climate Policy Radar empowers diverse stakeholders across the globe to demand better policies and practices that protect people and our planet. But there’s still a lot of work to do to reverse large-scale climate destruction and build future resilience. To learn more about her vision, we spoke with Dr. Nachmany about how AI can help and the values that drive her day-to-day work.

How did your journey inspire you to explore AI for humanity?

It’s worth saying upfront that humanity comes first. AI can, in certain circumstances and only when we exercise great care and humility, support humans working for humanity. So we start from the issues we want to solve, and not from the shiny new tools. This is how we began: from 2012, I was involved in helping parliamentarians find out what climate laws and policies already existed, so they could learn from each other and advance climate legislation. For several years, I led a research team at the London School of Economics, manually identifying and summarizing key documents. As this resource became essential to hundreds of thousands of users each year — including civil servants, researchers, investors, and activists — it became clear that we needed to scale our approach, with the goal of helping many different stakeholders research and understand more. This miraculously coincided with the emergence of AI, in particular Natural Language Processing (NLP), and I knew this was our chance to scale our impact. That is why I founded Climate Policy Radar in 2021.

Climate Policy Radar leads by example when it comes to climate collaboration. What unique tools and approaches have shown great promise for improving climate collaboration across industries, sectors, and geographies?

Climate progress depends on breaking down silos across languages, sectors, and disciplines so that people can access and use the knowledge they need. The simplest example is auto-translation. By removing language barriers, we allow people to access knowledge they previously couldn’t have. This is huge, especially for those who have fewer resources. But it’s also about pushing for openness, interoperability, and shared standards that make collaboration between organizations like ours the norm. We have no time to make big mistakes, and we need to push each other further as quickly and effectively as we can. We genuinely see other organizations as allies, not competitors — a deliberate rejection of the industry status quo. We focus on open science, explicitly sharing data, code, and methodology, we organize climate NLP tracks in academic conferences, and we also run a community of practice that offers abundant opportunities for learning and collaboration.

We know that marginalized groups, including Indigenous Peoples and women/girls, are too often disproportionately impacted by climate change. How is your organization using AI to help amplify their voices in climate legislation while safeguarding their rights to privacy, autonomy, and safety?

Marginalized communities are too often missing from the data that informs climate policy. We work to amplify their voices by making climate data more accessible, structured, and inclusive of their lived experiences. We actively consult with underrepresented groups to ensure that our methodology, tools, and datasets reflect diverse perspectives and needs. Since most training data skews toward the Global North, we focus on identifying and correcting biases to avoid reinforcing existing disparities. Our tools are free and open to everyone, but we don’t deal with personal or private data, so privacy and safety risks are less concerning in our case.

What core values drive your unique vision for impact in an AI-driven future?

Transparency. Sunshine is the best disinfectant. It’s important to share knowledge, methods, and limitations openly — to reduce bias, build trust, and give people agency by keeping them informed.

Humility. We say what we don’t know and can’t do. We also listen, adapt, and recognize AI’s limitations.

Mission appropriateness. Not every problem needs AI. Above all, AI products and platforms should serve impact and human values, not just efficiency. For example, are we using AI to optimize oil drilling, push fast fashion, or reinforce bias against the already marginalized; or are we using it to enhance learning, accelerate green investment, and improve grid efficiency?

Which visionary leaders, philosophies, or movements give you hope for a more human-centered AI future?

I think a lot about interbeing, the Zen Buddhist notion that all things are interconnected, interwoven, and mutually dependent. This means that how we conduct ourselves in our personal lives — how we relate to ourselves, our friends, partners, and children — is connected with how we build and run organizations and how we solve big global problems. Do we choose to prioritize collaboration or competition? Do we engage with compassion and inclusion, or suspicion and exclusion? It also means that climate and nature, food access and health, and human rights are all interconnected, and we can’t solve one issue without solving the others. I find this critical to thinking about AI — deciding what to use it for and how we should build it. Indeed, we are building our knowledge graph with all of these interconnected considerations in mind, rather than working in silos. My hope comes from seeing more and more leaders, allies, and colleagues speaking the same language.

What is your 7-word autobiography?

Systems reimaginer. Purposeful. Impatient. Sparkly. Straight-talker.