AI in Action: Google Research on Progress Towards the SDGs
AI in Action: Google Research on Progress Towards the Sustainable Development Goals. Will AI address humanity's biggest challenges.
There is no shortage of bold claims about what AI will do for humanity. Solve climate change, transform healthcare, fix education. The promises are easy to make. What is harder is showing the evidence. That is what makes a recent research brief from Google worth paying attention to.
The paper, titled “AI in Action: Accelerating Progress Towards the Sustainable Development Goals,” maps over 600 existing AI use cases against the UN SDGs. That is a 300% increase since 2018. You can read the full brief here.
What makes it interesting is not just the volume of use cases but where AI is having the most tangible impact. Health is a standout. Google’s AlphaFold project, which predicts protein structures with remarkable accuracy, is accelerating drug discovery in ways that would have seemed science fiction a decade ago. AI in medical imaging is improving disease detection. Predictive analytics is helping personalise treatment plans. These are not hypothetical applications. They are in use.
Education is another area where AI is making a difference, though the picture is more nuanced. Tools like Google’s Read Along app help children develop reading skills. Predictive analytics can flag at-risk students before they drop out. But the gap between what the technology can do and what most schools can actually implement remains wide, especially in the parts of the world that need it most.
Climate action is where the ambition meets the most obvious contradiction. AI is being used to forecast floods, optimise traffic light patterns to reduce emissions, and track greenhouse gas output through initiatives like Climate TRACE. All of this is genuinely useful. But AI itself has a growing environmental footprint, and the research brief acknowledges this directly. The energy demands of training and running large models are significant. Without cleaner infrastructure behind the technology, there is a real risk that the tool we are building to solve the problem makes the problem worse.
The brief also makes the case for responsible deployment, covering transparency, fairness, data privacy, and inclusivity. These are not afterthoughts. They are the conditions under which AI can actually deliver on its promises. Get them wrong and the technology entrenches existing inequalities rather than reducing them.
My take is that the potential is real but the jury is still out. We need more research and shared knowledge to properly understand AI’s impact across sustainability measures, including the SDGs. It still feels too early for a definitive verdict, but we do not have the luxury of waiting. A measured, evidence-led approach is the only responsible path.
One thing I would push harder on is the infrastructure question. Data centres should be run on clean energy as a bare minimum. Better still, the infrastructure itself should create benefit. Businesses like Deep Green are doing exactly this, capturing the waste heat from computing and repurposing it for social good. We need far more of that kind of thinking.
We work at the intersection of AI and sustainability. If your organisation is exploring how to use technology responsibly, take a look at our Sustainability & Circular Economy and Digital Product & AI work.
Further Reading
- Making Your Content AI-Friendly: A Guide to Getting Noticed by AI Assistants.
- AI Consultancy
- What are Sustainability Frameworks?
- The UN Sustainable Development Goals and Why They Are Important to Business.
- Artificial Intelligence and Sustainability. The Elephant in the Room.