Google Research

Today on the blog, read how a large-scale evaluation of a mammography AI system across multiple NHS screening services in the UK shows its potential to enhance cancer detection accuracy & reduce workload within complex double-reading workflows. More →goo.gle/4bMSjYZ

From AI tools helping radiologists detect 25% more interval cancers to agentic systems for clinical reasoning, today we announce new AI breakthroughs in healthcare that move innovation from the lab to real-world care. Learn more: goo.gle/41aCMMA

Check out #TheCheckUp tomorrow, where you'll hear from our own VP & Head of Google Research, @ymatias, and others ⬇️

Applications are open for the #GoogleOrgImpactChallenge: AI for Science, a global call for researchers + organizations using AI to achieve scientific breakthroughs for environmental sustainability, climate resilience & life sciences. Apply through Apr 17 →goo.gle/40oIXw0

Check out the full case study and see which models topped the leaderboard: goo.gle/4uyAK6k

LLMs are now unlocking a new level of scientific reasoning. We partnered with top experts to test 6 LLMs on high-temperature superconductivity, an open area of inquiry in condensed matter physics. Our case study found that curated, closed-system models were the clear winner, acting as true research partners by prioritizing high-quality, verified data over raw web volume.

@SALA7EDINN Appreciate the flag. We've corrected the map and reposted the announcement here to ensure accuracy. Thanks for the feedback!

@GoogleResearch Correct Moroccoan map 🇲🇦

@rtxtn Appreciate the flag. We've corrected the map and reposted the announcement here to ensure accuracy. Thanks for the feedback!

@GoogleResearch fix the Moroccan map @GoogleResearch

@Pixo14 @sundarpichai Appreciate the flag. We've corrected the map and reposted the announcement here to ensure accuracy. Thanks for the feedback!

@GoogleResearch @sundarpichai fix the moroccan map

@A380_Dice @GoogleArabia Appreciate the flag. We've corrected the map and reposted the announcement here to ensure accuracy. Thanks for the feedback!

@GoogleResearch outdated map of Morocco! @GoogleArabia

@BabaKhaliid Appreciate the flag. We've corrected the map and reposted the announcement here to ensure accuracy. Thanks for the feedback!

@GoogleResearch This is great thank you, we need as african to use this dataset to build applications for africans. Please make sure to correct the Morocco map thank you 🙏🏽

Note: We have updated this post announcement with a corrected map. We apologize for the previous error and appreciate the community feedback that helped us correct it.

The biggest barrier for AI applications in Africa isn't model complexity -- it's the scarcity of data for the 2000+ spoken languages there. We just released WAXAL. This open-access dataset delivers 2,400+ hours of high-quality speech data for 27 Sub-Saharan African languages, serving 100M+ speakers. Crucially, this community-rooted effort — led by African organizations — changes the roadmap for truly inclusive voice AI. Learn more and check out the WAXAL dataset here: goo.gle/4cxNHae

Introducing Groundsource: a new AI-powered methodology that transforms public information into a high-quality record of historical disaster data — starting with flash floods in urban areas. Using Groundsource we trained a new model that makes tangible progress towards predicting flash floods in urban areas up to 24 hours in advance. More details in 🧵⤵️

Learn how this work enabled the expansion of our global flood forecasting coverage to include flash flood forecasting in urban areas: goo.gle/4lrKiw7

Flash flood prediction models need historical data and model training that often doesn't exist. Our solution: Groundsource, a new AI-powered methodology that uses Gemini to transform 5M+ global reports into a precise dataset of 2.6M+ flood events. This provides a massive, open-source benchmark, scaling impact — especially in regions like Africa and Southeast Asia that lacked data. The same technique could be applied to other natural disasters. Read the technical breakdown on the Research Blog: goo.gle/46VcQbh

Breast cancer is one of the most common cancers in the world, and in the U.K. it affects 1 in 8 women. We partnered with Imperial College London and the NHS to see if AI can strengthen early detection efforts. The result: Our experimental research AI system identified 25% of the “interval cancers” that typically slip through traditional screenings.