Google A.I. Can Detect Risk of Heart Disease Through Eye Scan The search engine announced the news on Monday. Together with Verily, Google's health-tech branch, they have developed an A.I. that can determine if a person has high blood pressure or is at risk of a stroke or heart attack, just by scanning their eye. Researchers were able to "train" the A.I. by showing it thousands of images of scanned retinas, allowing it to recognize when a person is at risk for heart problems. Right now, the algorithm is

This website creates a fake face on each refresh

Every one of the people in the images below is fake.

Philip Wang, a software engineer at Uber, made a site for generating fake faces. uses an algorithm that’s trained with a database of real portraits. The algorithm then uses a generative adversarial network, a type of neural network, to create a new fake face each time you refresh the site.

“Most people do not understand how good AIs will be at synthesizing images in the future,” Wang told Motherboard. He wrote on Facebook he set up the site to raise awareness of the technology.

The software behind the site isn’t incredibly obscure, either. The site uses a GPU (graphics card) running code published by Nvidia researchers.

» Douglas Rain, voice of HAL from ‘2001: A Space Odyssey,’ dead at 90

According to a writeup of the tech, Nvidia’s algorithm trains the AI using progressively higher-resolution images.

This GIF, courtesy of machine learning engineer Sarah Wolf, shows the process in action.

Photo: Sarah Wolfe

The technology could have profound implications for the future of photographic evidence. As The Verge put it, “ is just the polite introduction to this new technology. The rude awakening comes later.”

Support real journalism. Support local journalism. Subscribe to The Atlanta Journal-Constitution today. See offers.

Your subscription to the Atlanta Journal-Constitution funds in-depth reporting and investigations that keep you informed. Thank you for supporting real journalism.