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 foundational research that leads directly to real-world applications and solutions. These solutions not only benefit millions around the world, but they also uncover more important problems to solve.
The magic cycle is accelerating significantly, powered by stronger models, and agentic tools, like the AI co-scientist and AI-based expert-level empirical software. And it applies across many disciplines and domains.
Research is our chance — and imperative — to improve day-to-day lives, and address societal challenges and opportunities, and it means that research and innovation is never “done.”
Here’s how Google Research’s approach is helping 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, requiring tailored treatments based on their specific mutations. What if we could sequence the genomes of these cancerous cells more accurately, precisely spotting the particular variants that turned them into cancer?
That’s what led to DeepSomatic, our new AI-powered tool that helps scientists and doctors spot genetic variants in cancer cells. Our partners at Children’s Mercy in Kansas City used DeepSomatic to identify 10 new genetic variants in samples of childhood leukemia that were missed by previous techniques. If they can pinpoint how and why a particular form of cancer is affecting a patient, they may be able to develop personalized cures.
Remarkably, DeepSomatic can also generalize to cancers it hasn’t seen before. For example, without any training on the brain cancer glioblastoma, DeepSomatic was able to pinpoint which genetic variants cause it. This suggests it could work even on rare or new types of cancer — a big milestone that marks 10 years of genomics research at Google.
In collaboration with Yale and Google DeepMind, we also introduced Cell2Sentence-Scale 27B, a new 27-billion-parameter AI model based on Gemma that understands the language of individual cells. It generated a novel hypothesis for cancer therapy, which we validated in living cells, finding a drug combo that made cancer cells significantly more visible to the immune system in a lab setting. It’s a powerful new way we’re using AI to help fight cancer.
2. Towards better medicines and materials with quantum computing
Designing better medicines and materials — like a more effective 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 can’t capture all the nuances of how molecules behave in the wild. That’s because at this tiny scale, particles don’t behave “classically.” Instead, they obey quantum mechanics: they can be in superposition, where they’re not in one simple state, but instead “smeared” across a range of possibilities; and they can be entangled, where multiple atoms can behave in lockstep with each other instead of independently.
This is one of the most compelling reasons 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 computations that are very useful for describing how molecules behave with full precision. This is the world’s first algorithm that points towards eventual practical applications of quantum computing, like designing better materials, better medicines and more. We’ve begun exploring its potential in experiments in partnership with researchers at the University of California, Berkeley.
3. Understanding 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 pulling it all together. For example, if we want to predict which communities are most vulnerable and what infrastructure is at risk with an impending storm, it’s not enough to know simply 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 knit all of that information and predictive power together. Questions that are currently impossible to answer because they’re too complex and draw on too many disparate geospatial resources will become possible to tackle. And this, in turn, will prompt 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 uses and new research, uncovering even deeper insights about how we can live well on this planet.
Those are just 3 areas — out of dozens! — where Google Research is making foundational breakthroughs and then demonstrating how they can be scaled up for real, tangible impact for people. These breakthroughs don’t happen in silos: we’re driven by the conviction that breakthroughs in one area of scientific discovery can help us in another — be that with better quantum data to accelerate AI discoveries, or connecting geospatial data with public health insights. That’s how we’re paving the path to the future, all founded in research that can make reality better for people.
