At a time when fears of the impacts of artificial intelligence are on the rise, Google LLC said today that it will give away $25 million toward projects that use its artificial intelligence technologies for “social good.”

The announcement, made at a demonstration area in the company’s Sunnyvale, California facilities for some of Google’s and other organizations’ AI-driven social-good projects, will funnel money to nonprofits, universities and other organizations that will participate in an AI Global Impact Challenge. It’s part of a larger new initiative called AI for Social Good also announced today.

Google and other companies ranging from Amazon.com Inc. to Facebook Inc., has come under increasing fire in recent years for enabling artificial intelligence to be used for what some consider less-than-socially conscious purposes, from recognizing people in military drone images to eliminating jobs.

The projects and the new grant might be seen as a public-relations move to ease criticism both by the public and by Google’s own employees, whose opposition to a military project caused the company to forgo a renewal of it and also to come up with a set of principles for using AI. Jeff Dean, Google’s head of AI, insisted the latest initiatives are “not really related.”

“There’s never been a better time to be working on artificial intelligence,” Dean said, citing breakthroughs in areas such as deep learning neural networks that have led to much-improved image and speech recognition, self-driving cars and other wonders. At the same time, he acknowledged that “we’re all grappling with new questions about AI.”

In any case, even as there’s no doubt that AI has touched off societal worries about its impact, it’s just as clear that no small number of Googlers themselves are hoping to apply AI to more socially conscious or at least unambiguously positive uses.

Google AI chief Jeff Dean (Photo: Robert Hof/SiliconANGLE)

Among the projects underway are one with the National Oceanic and Atmospheric Administration on whale bioacoustics that uses a high-frequency acoustic recording package (pictured) to detect cetacean sounds underwater and a neural network to detect and differentiate humpbacks and other species of whales. The aim is to let shipping companies know where whales are located to avoid collisions.

Another project aims to improve flood forecasting, using Google machine learning to simulate floods on a river in Hyderabad, India, with more accuracy than existing methods. Pete Giencke, a product manager on Google’s Crisis Response team within its Google.org philanthropic organization, told SiliconANGLE that the company hopes to leverage the information to create not only alerts when people search for flood information, but even arrange for locals to spread the word by bicycle or other means.

Other AI-driven projects on which Google is working with other organizations are aimed at identifying and reducing child trafficking and helping scientists find a new planet in other solar systems. Internally, Google is working on a Lookout app that provides a continuous audio stream describing text and objects for blind people.

Google isn’t alone in addressing social and ethical issues raised by the advance of AI into many industries and corners of everyday life. Microsoft Corp. already has its own AI for Good program to which it has committed $115 in grants for environmental, humanitarian and accessibility projects.

Google.org will announce the winners of its challenge at the Google I/O software developer conference next spring.

Photo: NOAA Fisheries