Google’s A.I. can now detect breast cancer more accurately than doctors can

By Allison Matyus Published April 25, 2023

Google’s artificial intelligence technology is able to spot signs of breast cancer in women with more accuracy than doctors, according to a new study.

Recommended Videos

The study, published on Wednesday, January 1, in the scientific journal Nature, found that by using A.I. technology, there was a reduction in false positives and false negatives when it came to diagnosing forms of breast cancer.

A.I. technology was used to look the mammograms from more than 15,000 women in the United States, and over 76,000 women in the United Kingdom. The program was able to reduce false positive by 5.7% for women in the U.S. and 1.2% for those in the U.K. False negatives were reduced 9.4% in the U.S. and by 2.7% in the U.K.

The advanced A.I. system proved to be more accurate than human experts with knowledge of a patient’s history, even if doctors did a second reading of mammogram results, according to the study.

“The performance of even the best clinicians leaves room for improvement,” the study reads. “A.I. may be uniquely poised to help with this challenge.”

The study was a collaboration between Google Health, Northwestern University, Cancer Research U.K. Imperial Centre, and Royal Surrey County Hospital.

“Looking forward to future applications, there are some promising signs that the model could potentially increase the accuracy and efficiency of screening programs, as well as reduce wait times and stress for patients,” wrote Google Health’s Technical Lead, Shravya Shetty and Product Manager, Daniel Tse, in a blog post announcing the initial findings.

The use of A.I. technology to better detect screenings could be groundbreaking, considering that one in eight U.S. women will develop breast cancer in their lifetime, according to the American Cancer Society.

Aside from health applications, Google’s A.I. technology is also being applied for identifying different species of animals in a new wildlife conservation program that aims to better protect and monitor animals throughout the world.

The ultimate goal for the data is to identify different species of animals faster. According to the blog, human experts can look through 300 to about 1,000 images per hour, but Google’s A.I. technology can analyze 3.6 million photos an hour and automatically classify an animal.

Former Digital Trends Contributor

Allison Matyus is a general news reporter at Digital Trends. She covers any and all tech news, including issues around social…

Topics

Emerging Tech

OpenAI admits it needs to rethink what happens when AI goes rogue

As increasingly capable AI agents move beyond controlled tests, OpenAI is confronting a difficult question: When does strange model behavior become an incident the public deserves to know about?

OpenAI has spent plenty of time explaining how it plans to stop increasingly capable AI agents from doing things they shouldn't. Now, the company says it needs to get better at telling everyone when those things have already happened. The admission follows reports of another previously undisclosed incident involving OpenAI's AI agents, this time affecting a German-language programming wiki. According to Reuters, agents made more than 15,000 unauthorized edits to DseWiki, using the site to communicate and share ways to bypass restrictions, cheat on tasks, and avoid detection.

OpenAI has now acknowledged what it calls the "wiki incident" and says the episode exposed a larger problem with how AI companies disclose unexpected model behavior. The company says it is developing a framework for deciding when and how to make incidents involving misaligned AI public.

Emerging Tech

AI has a safety problem nobody is ready for

AI safety efforts are ramping up, but their protections don’t work equally well everywhere. In developing countries, language gaps and cultural blind spots can turn everyday AI mistakes into serious real-world consequences.

AI companies are spending an enormous amount of time worrying about what happens when their models become too capable. OpenAI even took the unusual step of temporarily pausing training on a model last month over safety concerns, as the industry grapples with risks ranging from autonomous behavior to increasingly sophisticated cyber capabilities.

But there’s another AI safety problem that is much easier to overlook: the protections already being built into these systems don’t necessarily work equally well for everyone. A new report from Rest of World highlights how AI safety efforts remain heavily centered around the needs of wealthier, English-speaking countries. That can leave people in parts of Asia, Africa, and other developing regions dealing with problems as basic and potentially dangerous as a chatbot misunderstanding their language.

Branded Content

The IFA 2026 Publisher Award Winners Bring Fresh Ideas to Everyday Tech

From Smarter Entry to Screen-Free Wellness, These IFA 2026 Winners Do Things Differently

IFA 2026 Publisher Awards

IFA is the kind of event where almost every corner promises the next big thing in tech. Some ideas are ambitious, others are surprisingly simple, and a few make you wonder why nobody thought of them sooner. But among hundreds of new products, the most memorable ones usually have something more going for them than novelty alone.

This year’s IFA Publisher Award winners are a good example. They span everything from smart entry and personal wellness to coffee, floor care, and big-screen entertainment. What connects them is how they take technology in interesting new directions while keeping its purpose grounded in how people actually use it.