The Good and the Bad of AI and Climate Change

Artificial intelligence (AI), information and communication technology, and other tools are increasingly popular in the fight against climate change. As countries strive to achieve net-zero, technological innovations in solar energy and predictive AI for weather modeling are gaining popularity.

Tech companies are among the most significant carbon polluters. The production of silicon chips or semiconductors, found in many gadgets and products today, requires enormous energy. This accounts for the majority of carbon emissions from electrical devices. Many tech companies have declared net-zero policies, but these policies are far from achieved. The advancements of AI and technology are also exacerbating global inequality since most of AI’s social and economic benefits are only enjoyed by a select few.

In our podcast’The Contrary by IDR’, we sat with Jim Fruchterman from Tech Matters and Trisha Raymond of Observer Research Foundation. We discussed the role of technology in climate solutions.

The following is an edited transcription that gives a summary of the guest’s perspectives on the show.

AI can accelerate climate change

Jim : Currently, AI is used mainly by academics to create climate models. They use it to create climate models to better understand the climate crisis. They also use AI to analyze the current situation and to forecast the future. I believe people are using technology in a variety of ways and, more broadly, to combat climate change. You can see this most in the energy sector. Energy is the biggest area where people are trying to innovate. Solar panels are also [available], and they’re not necessarily considered a high-tech product. It is a large part of wind energy.

Trisha :Technology like AI is capable of a lot. It can help us track and predict climate change, as well as mitigate it. AI could be used to create new synthetic materials that would replace those we currently use. We can create new materials with better heat resistance, or that help us to store solar energy. A smart grid is also a concept that has a lot to offer. India also has its own National Smart Grid Mission, where AI could be used to detect patterns of electricity usage and changes over time.

AI could be used to create new synthetic materials that would replace those we use today. | Picture courtesy: CGIAR Climate / CC BY

Unregulated AI use can lead to serious consequences

  1. AI models emit their own carbon footprint

Trisha : We talk about climate tech in a way that makes us think it is a solution to all of our problems. These applications come at a price… A way to look at this issue is to realize that training an AI model takes a great deal of energy. A study has been done to try and quantify the carbon emission [of AI]. It says that the training of a single model for natural language processing emits as much CO2 as an entire car over its lifetime. It’s quite a bit.

  1. Data colonialism is a result of the existing business model in tech

Trisha :The economic and social benefits of AI accrue to a few privileged countries. The fact that their citizens’ data is being exported and that services and products are coming in is a major concern for developing and underdeveloped nations. They’ve now become both data suppliers and product purchasers. This concept is called data colonialism. Sub-Saharan Africa and Latin America, the Caribbean, South, Central Asia, and South Asia are lagging behind in AI use and development in terms of funding, start-ups, and other skills.

Jim :[Technology companies] are stealing your data to become billionaires. We need to use data and AI built on that data more for the benefit of the global society, the local communities, and patient states. Nithya and I have published a number of pieces about the need to decolonize the data. These are not aimed at companies because they process clinical data in total, but rather at government and nonprofits to help them understand that, while empowering communities, they cannot do so at the cost of extracting data and using it against that community.

  1. The tech industry isn’t doing nearly enough to offset its emissions

Jim Tech firms are powerful and kill legislation that is against their interests. We haven’t seen any privacy or data usage laws in India, for example, for a long time. And tech companies have good social arms. The social good arm’s idea is to decide how we will spend, say, 0.1 percent or 1 percent of profits. Let’s ignore what we do for the other 95% of the time. We’re 10 to 15 years behind in our social work, so we use these models very little.

Trisha : [To fight emissions caused by technology] Many major tech giants–, Microsoft, Alphabet, and Facebook- have all announced policies and initiatives. There’s still much to do. The Carbon Offsets System is often used to support these net-zero projects. Amazon, for example, can declare a net environmental impact of zero if it plants a certain number of trees elsewhere in the world. In reality, things don’t work that way. There is no simple equivalent. Amazon should fundamentally change how it operates and be more transparent about its environmental impact.

What is needed to be changed?

  1. AI in communities

Jim : I would like to see AI used to assist the local farmer and community leader to better understand the situation to make informed decisions based on their priorities and context. How can you make things better at the local level with this landscape focus? Local leaders also don’t give much thought to climate. The local leaders are more concerned with how to boost the income of farmers, despite having less water. They are aware of the changes but want to concentrate on local issues. This landscape approach aims to give local leaders and producers, as well as smallholder farmers and producers, more power. It is led by traditional aid organizations and biodiversity and conservation organizations, who know they will only achieve their goals with local communities making climate-smart choices.



Author: admin

Leave a Reply

Your email address will not be published. Required fields are marked *