Over the past two weeks, we’ve seen a flurry of new results from Google Research, from genomics to quantum computing to geospatial understanding.
These breakthroughs exemplify what I call the “magic cycle” of research: addressing global challenges and opportunities with basic research that leads directly to real-world applications and solutions. These solutions not only benefit millions around the world, but they also reveal more important problems to solve.
The magic cycle is accelerating significantly, driven by stronger models and agent tools, such as the AI co-researcher and expert-level AI-based empirical software. And this applies across many disciplines and domains.
Research is our chance – and imperative – to improve everyday life and address societal challenges and opportunities, and this means that research and innovation are never “finished”.
Here’s how Google Research’s approach helps solve challenges in three areas that affect so many people:
1. Fighting cancer with AI
Childhood leukemias and many other cancers have incredibly complex genetic signatures that require tailored treatments based on their specific mutations. What if we could sequence the genomes of these cancer cells more precisely, pinpointing the particular variants that made them cancerous?
That’s what led to DeepSomatic, our new AI-powered tool that helps scientists and doctors find genetic variants in cancer cells. Our partners at Children’s Mercy in Kansas City used DeepSomatic to identify 10 new genetic variants in childhood leukemia samples that were missed by previous techniques. If they can figure out how and why a particular type of cancer affects a patient, they may be able to develop personalized cures.
Remarkably, DeepSomatic can also generalize to cancers it has not seen before. For example, DeepSomatic was able to figure out which genetic variants cause it without training in the brain cancer glioblastoma. This suggests it may work even on rare or new types of cancer – a major milestone marking 10 years of genomics research at Google.
In collaboration with Yale and Google DeepMind, we also introduced Cell2Sentence-Scale 27B, a new AI model with 27 billion parameters based on Gemma that understands the language of individual cells. It generated a new hypothesis for cancer treatment, which we validated in living cells, finding a combination of drugs that made cancer cells significantly more visible to the immune system in a laboratory setting. It’s a powerful new way we’re using artificial intelligence to help fight cancer.
2. Towards better medicines and materials with quantum computing
Designing better medicines and materials—like a more efficient battery—requires understanding the exact behavior of atoms and molecules. But today’s most powerful classical computers struggle to model these nuances because they rely on approximations and the strict binary language of 0s and 1s, and even the world’s most powerful supercomputer cannot capture all the nuances of how molecules behave in nature. This is because particles on this small scale do not behave “classically”. Instead, they obey quantum mechanics: they can be in superposition, where they are not in one simple state, but instead “smeared” over a range of possibilities; and they can be entangled, where multiple atoms can behave in lock step with each other rather than independently.
This is one of the most compelling reasons why Google Research is building a quantum computer – it “speaks quantum” in a way that no classical computer can, and can model exactly how nature really works at a subatomic level. Our new Quantum Echoes algorithm shows how much faster our Willow chip can be on calculations that are very useful for describing how molecules behave with full precision. This is the world’s first algorithm that points toward possible practical applications of quantum computers, like designing better materials, better medicine, and more. We have begun to explore its potential in experiments in collaboration with researchers at the University of California, Berkeley.
3. Understand the Earth
The hardest and most important questions in planetary science and crisis response are never about just one kind of geospatial information—they’re about bringing it all together. For example, if we want to predict which communities are most vulnerable and which infrastructure is at risk with an impending storm, it is not enough to simply know when the storm will hit or where the buildings are. We need the whole picture: the storm’s path and severity, population density, transportation patterns, and predicted impact on vulnerable infrastructure. This comprehensive view requires synthesizing many types of geospatial data and many models that predict different aspects of the planet—all at once.
That’s why we’re developing Earth AI – to bring all that information and predictive power together. Questions that are currently impossible to answer because they are too complex and draw on too many different geospatial resources will become possible to tackle. And this, in turn, will spark new research—new collection of useful data about Earth, new kinds of sensors, and new uses of AI to model sophisticated interconnected patterns across the planet. This multi-year effort will cycle continuously between new real-world applications and new research, revealing even deeper insights into how we can live well on this planet.
That’s just 3 areas – out of dozens! — where Google Research makes fundamental breakthroughs and then demonstrates how they can be scaled up to real, tangible impact for people. These breakthroughs don’t happen in silos: we’re driven by the belief that breakthroughs in one area of scientific discovery can help us in another—be it with better quantum data to accelerate AI discoveries or connecting geospatial data with public health insights. This is how we pave the way for the future, all based on research that can make reality better for people.