Using AI, Microsoft and UNFCCC take aim at emissions

An AI-powered platform will analyze carbon data for Paris Agreement signatories.

November 30, 2023
Line of data servers

Photo credit: Alberto Ortega / Europa Press via Getty Images

The good news: we are likely on track to meet the COP28 hosts’ goal of tripling renewables by 2030. 

The bad: There is general agreement that we are not on track to achieve the Paris Agreement’s goal of limiting warming to 1.5 degrees Celsius, in large part because we are not on track to reach net zero emissions by 2050.

In fact, a new report from McKinsey found that  the current trajectory of emissions means we aren’t even on track to meet net zero by the end of the century. And according to a recent United Nations report, by 2030 we’re already likely to be behind on interim targets, to the tune of between 20.3 and 23.9 gigatons. To get back on track, global emissions must peak within the next two years.

Chart credit: McKinsey & Company

Against the backdrop of that narrow window, both the public and private sectors are increasingly turning to new technologies, in the hopes that they might unlock the path to net zero.

On the eve of COP28, the UN Framework Convention on Climate Change announced a partnership with Microsoft to bring artificial intelligence to the emissions challenge.

Through that partnership, Microsoft will back an AI-powered platform for measuring and analyzing emissions reduction progress, providing a layer of digital support for the UNFCCC’s Enhanced Transparency Framework.

Tracking global carbon dioxide emissions is notoriously difficult, given the immense quantities of data that must be validated and analyzed. Then of course, there’s the challenge of varying reporting formats and general inconsistency between member nations. And, due to the complexity of the accounting tasks, there are often reporting delays, which have caused gaps between data and policy.

Accordingly, Microsoft global operations and go-to-market leader of energy and resources Hanna Grene said at an October Latitude Media event that data summarization is one of the key applications for AI and machine learning. 

Microsoft’s forthcoming collaboration with the UNFCCC will allow for the “advanced analysis of global climate data,” the company said, and will equip countries with tools to report and validate emissions for industries like  transportation, agriculture, and manufacturing.

Microsoft is also developing tools for methane detection, Grene told the audience last month. Those tools use AI to learn from the patterns of methane leaks in the oil and gas and utility spaces, she said, and will eventually detect leaks before they happen.

Artificial intelligence is already being applied to many elements of the energy transition, from boosting renewables production and predicting energy prices to building a responsive grid. But for the most part, adoption of new AI tools, particularly for the energy sector, is happening slowly.

Microsoft and the UNFCCC didn’t specify the development and implementation timeline of their new platform and accompanying climate data hub, but given the increasingly tight timeline to get on track for Paris Agreement targets, the partnership is aimed at pushing both countries and the UN itself towards faster and more accurate reporting.

“Simply put, you can’t fix what you can’t measure,” Microsoft vice chair and president Brad Smith said in a statement. “These new AI and data tools will allow nations to measure emissions far better than they can today.”

2023 in Review: A Report on the Booming US Clean Energy Tax Credit Market

A wide range of companies can now invest directly in domestic wind, solar, bioenergy, EV, and manufacturing projects. Learn more in this free event, supported by Crux.

January 31, 2024 1:00 PM
Register HereWatch Now
No items found.