Technology is not just about improving business efficiency, customer experience and overall security. It is also about making a large human impact.

Scientists are using artificial intelligence and machine learning to study, for example, space weather and its major impacts on Earth’s infrastructure, oceans and their relationship to climate change, as well as global food supply and a way to improve it.

“Machine learning should be thought of as an extension of what a human is really good at; we can train machines to do this on a much grander scale,” said Sebastien de Halleux (pictured, left), chief operating officer of Saildrone Inc., which designs, manufactures and operates a global fleet of wind and solar-powered ocean drones monitoring the state of the planet in real time.

“If you partner with people who really have strong AI knowledge, you can use your knowledge of science to get to the really important issues,” stated Janet Kozyra (right), space weather scientist at the National Aeronautics and Space Administration.

Halleux, Kozyra and Henry Sztul (center), senior vice president of science and technology at Bowery Farming Inc., spoke with John Furrier, host of theCUBE, SiliconANGLE Media’s mobile livestreaming studio, during the AWS re:Invent conference in Las Vegas. They discussed how they’re shaping technology into scientific projects to make human life better.

How to learn from space, oceans and farms

Technology can help answer scientific questions in ways never before possible, according to Sztul, who works on a project that aims to revolutionize agriculture with AI. At Bowery Farming, he is building a network of large-scale indoor warehouse farms where it is possible to grow all sort of produce 365 days a year using zero pesticides, hydroponic systems, and LED technology.

“The core of it is some technology we call the Bowery Operating System, which is how we leverage software, hardware and AI to operate and learn from our farms,” he explained. “We have eyes on every single crop that grows in our facilities, and so we process those, learn from that data, and funnel that back into the system.”

AI and ML are also powerful tools in Saildrone’s data collection platform. It uses wind-powered robots to study the 70% of the planet that is really scarce in data: the oceans.

“We measure things like biomass, which is how many fish there are in the ocean. We measure the input of energy which impacts weather and climate, we map the seabed, and we do all kind of different tasks which are very, very expensive to do if you use ships,” Halleux stated.

Data collected from the oceans will be useful for understanding and addressing climate change, for example. “The ice is melting, the Gulf Stream is changing, El Niño is wreaking havoc, but we just do not understand this because we just don’t have the data in situ,” Halleux said. “We have to do better; we have to transform this into a big-data problem.”

This use of technology does not come without obstacles. While studying space superstorms, NASA found a big-data problem, according to Kozyra, who is partnering with Amazon Web Services Inc. in this study. This is because storms of that magnitude are very rare, so it is difficult to find enough data to train AI.

“We’ve decided to take the approach that these superstorms are like anomalies in the normal weather patterns,” she said. “So we’re trying to use the kind of AI that you use to detect anomalies like people who are trying to break into, to do bank fraud, or do web server attacks.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of AWS re:Invent:

Photo: SiliconANGLE