If you care about the environment, think twice about using AI.
Generative artificial intelligence uses 30 times more energy than a traditional search engine, warns researcher Sascha Lucioni, who is on a mission to raise awareness about the environmental impact of this new technology.
The Russian-born Canadian computer scientist, recognized by the American magazine Time as one of the 100 most influential people in the world of AI in 2024, has been trying to measure the emissions of programs like ChatGPT or MidJourney for several years.
“I find it particularly disappointing that generative AI is used to search the internet,” the researcher lamented, speaking to AFP on the sidelines of the All In Artificial Intelligence conference in Montreal.
The language models on which the programs are based require enormous computing capabilities to train on billions of data points, which in turn requires powerful servers.
Energy is then used to respond to each individual user’s requests.
She explains that rather than simply extracting information, “like a search engine does to find the capital of a country,” AI programs “generate new information,” making the whole process “much more energy-intensive.”
According to the International Energy Agency, the combined AI and cryptocurrency sectors will consume around 460 terawatt hours of electricity in 2022 – representing two percent of total global generation.
energy efficiency
A leading researcher on the impact of AI on climate, Lucioni participated in the creation of a tool for developers in 2020 to measure the carbon footprint of running a piece of code. “CodeCarbon” has since been downloaded more than a million times.
The head of climate strategy at startup Hugging Face, a platform for sharing open-access AI models, she is now working on building a certification system for algorithms.
Similar to the US Environmental Protection Agency’s program, which awards points based on the energy consumption of electronic devices and appliances, this will make it possible to know the energy consumption of an AI product to encourage users and developers to “make better decisions.”
“We don’t take into account water or scarce materials, but at least we know that for a specific task, we can measure energy efficiency and say this model got an A+, and that model got a D,” she admits.
Transparency
To develop her tool, Lucioni is experimenting with generative AI models that are accessible to everyone, or open source, but she also wants to try it on the commercial models of Google or ChatGPT-maker OpenAI, which are reluctant to agree.
Although Microsoft and Google have committed to achieving carbon neutrality by the end of the decade, the US technology giants see their greenhouse gas emissions increase drastically in 2023 due to AI: a 48 percent increase for Google compared to 2019 and a 29 percent increase for Microsoft compared to 2020.
“We are accelerating the climate crisis,” Lucioni said, calling for greater transparency from technology companies.
The solution, he says, could come from governments that are currently “flying blind” without knowing “what’s in the data sets or how the algorithms are trained.”
“Once there is transparency, we can start legislating.”
‘Energy Restraint’
According to Lucioni, “It’s also important to explain to people what generative AI can and cannot do, and what its costs will be.”
In their latest study, researchers demonstrated that creating a high-definition image using artificial intelligence consumes the same amount of energy as fully recharging your cell phone battery.
At a time when more and more companies want to integrate technology into our lives – with conversational bots and connected devices, or in online searches – Lucioni advocates “energy restraint.”
He stressed that the idea here is not to oppose AI, but to choose the right tools and use them judiciously.
(Except for the headline, this story has not been edited by NDTV staff and is published from a syndicated feed.)
