Sannu’s Reviews > Taming the Sun: Innovations to Harness Solar Energy and Power the Planet > Status Update

By using a machine-learning algorithm to predict, based on past records and current weather data, when such renewable generators as solar panels will flood the grid with power, Microsoft can adjust its data centers’ operations to consume that surplus solar power.

Softbank won a major contract to build a utility-scale PV project in the Indian state of Andhra Pradesh. Within a year, Raman triumphantly sent me a Google Earth satellite photo of three adjacent project sites. Two were barren—waiting to be built on by slow-moving local firms—but Softbank’s site already brimmed with 350 MW of installed solar panels.

Down the road, the next step is to expand high-quality data collection to developing countries, where skittish investors need hard data to get comfortable with plowing their capital into markets having both high growth potential and high perceived risks.

Because the more solar that is installed, the less that the electricity it generates in the middle of the day is needed. On top of this, higher levels of solar power strain the rest of the electricity system, imposing costs or, equivalently, reducing solar’s value.

By mid-century, India could vault above China and the United States to become the world’s biggest emitter of greenhouse gases. In response, the Modi government’s 100-GW solar power target is the most ambitious part, by far, of its climate pledge under the 2015 Paris Agreement.

As the population grows, farmers move to the cities, and a swelling middle class chases the energy-intensive trappings of a modern lifestyle, India could use four times as much power by 2040 as it does today.