How AI causes serious environmental problems (but might also provide solutions) | DW Business

Summary

This script explores the dual-edged role of AI in environmental conservation. While AI excels at analyzing vast data sets to optimize energy use and predict climate patterns, its resource-intensive nature poses environmental challenges. Data centers' high energy and water consumption, especially in drought-stricken areas, raise sustainability concerns. Companies like Microsoft are investing in carbon offsets, but questions remain about the effectiveness and potential greenwashing. The script calls for sustainable AI development, community involvement in governance, and stricter regulations to mitigate AI's environmental impact.

Takeaways

• 🧠 AI's Potential in Climate Solutions: AI can analyze vast amounts of data quickly, which is invaluable for optimizing energy use and making decisions to reduce waste.
• 🏢 Energy Efficiency in Buildings: Companies like Adian use AI to help facility managers run commercial buildings more efficiently, potentially saving up to 40% of energy and CO2 emissions.
• 🔍 AI's Speed and Analysis: AI's ability to read and analyze data rapidly is a significant advantage over human researchers, allowing for swift responses to environmental issues.
• 🌡️ The Environmental Impact of AI: The operation of AI, particularly in data centers, consumes substantial resources like water and electricity, contributing to environmental challenges.
• 💧 Water Usage Concerns: Large-scale AI operations, such as data centers, can strain local water resources, as seen in the controversy over Google's data center in Chile.
• 🔥 Energy Consumption and Emissions: Training and running AI models require significant energy, which often comes from fossil fuels, contributing to CO2 emissions.
• 🌐 Global Data Center Growth: The expansion of data centers worldwide is expected to double energy consumption by 2026, highlighting the growing environmental impact of AI.
• 🌿 Sustainable AI Initiatives: Some companies, like Microsoft, are investing in carbon offset projects to counterbalance the environmental impact of their AI operations.
• 🏛️ Regulatory Efforts: Governments and regulatory bodies are beginning to address the environmental impact of AI through new rules and legislation, like the EU AI Act.
• 🌳 Green Alternatives: Tech companies are exploring renewable energy sources for data centers, such as solar and wind power, to reduce their carbon footprint.
• 🌍 Community Involvement: The script emphasizes the importance of involving local communities in AI governance to ensure sustainable and considerate technological development.

Q & A

What is the potential of AI in addressing the climate crisis?

AI has the potential to analyze large amounts of data quickly, which can help in optimizing energy use in buildings, mapping Antarctic icebergs, tracking deforestation, predicting weather patterns, and suggesting efficient waste management systems.

How does AI technology contribute to the problem of climate change?

AI, particularly through data centers, consumes significant resources like water and electricity. The energy used often comes from burning fossil fuels, which release greenhouse gases contributing to climate change.

What is the role of Adian in managing large commercial buildings?

Adian is a startup that helps facility managers run large commercial buildings efficiently by using AI to analyze data from the building's technical systems and suggesting optimizations to save up to 40% of energy and CO2 emissions.

How does Adian's AI system assist with error resolution in building systems?

Adian's system allows facility managers to query error codes directly, and the AI, using the building's historical data, can provide specific instructions on how to resolve the issue, making the process faster and more efficient.

What are some of the environmental impacts of data centers?

Data centers require large amounts of water for cooling and consume significant electricity, which often comes from non-renewable sources. This can lead to conflicts over water resources and contribute to greenhouse gas emissions.

Why were Google's plans for a data center in Chile put on hold?

The plans were halted due to local concerns that the data center would consume a large amount of water, 168 liters per second, in an area already facing a severe drought, potentially impacting the local community's access to drinking water.

What is the embodied carbon of training a single AI model?

Training a single AI model produces more than five times the amount of carbon dioxide emissions generated from a car in its lifetime, including emissions from manufacturing the car and its fuel consumption.

What steps is Microsoft taking to address its carbon footprint with AI expansion?

Microsoft aims to be carbon negative by 2030 and is investing in carbon dioxide removal credits from a biochar facility that turns forestry waste into a form of charcoal, which is then buried to enrich soil and sequester CO2.

What is the EU AI Act, and how does it relate to environmental impact?

The EU AI Act is a set of rules that will impact businesses using AI, emphasizing the development and use of AI systems in a sustainable and environmentally friendly manner, although it does not provide specific guidelines on what this entails.

What is the Climate Neutral Data Center Pact, and what commitment does it entail?

The Climate Neutral Data Center Pact is an agreement signed by around 100 data center operators, promising to operate climate-neutral data centers by 2030, aiming to reduce the environmental impact of their operations.

How are some governments and companies addressing the strain of data centers on power grids and resources?

Governments like Singapore have enacted moratoriums on new data center constructions, while companies are experimenting with green alternatives such as using solar energy and wind power to run their data centers.

Outlines

00:00

🤖 AI's Role in Climate Crisis: Savior or Contributor?

The script explores the dual nature of AI in the context of environmental conservation. It acknowledges AI's potential to analyze vast amounts of data quickly, which is instrumental in optimizing energy use in buildings and suggesting efficient waste management systems. However, it also highlights the significant energy and water consumption of data centers, which can lead to environmental issues like the depletion of local water sources. The episode discusses how AI can help in energy optimization and the challenges it poses due to its resource-intensive nature.

05:00

💧 The Environmental Impact of AI and Data Centers

This paragraph delves into the environmental implications of AI, particularly focusing on the resource demands of data centers. It discusses the controversy surrounding Google's data center in Chile, which faced opposition due to its high water consumption in a drought-stricken area. The paragraph also addresses the broader issue of AI's carbon footprint, from the energy used in training models to the ongoing electricity needs of data centers. It raises questions about the sustainability of AI and the need for corporate responsibility in mitigating its environmental impact.

10:01

🌱 Balancing AI Advancement with Environmental Sustainability

The final paragraph discusses the efforts of tech companies to balance the advancement of AI with environmental sustainability. It mentions initiatives like Microsoft's aim to be carbon negative and the use of biochar to sequester carbon dioxide. The paragraph also touches on the challenges of regulation, with the EU's AI Act and the Climate Neutral Data Center Pact, and the need for more comprehensive environmental considerations in AI development. It concludes by emphasizing the importance of community involvement in AI governance and the potential for AI to contribute positively to climate change mitigation, while also recognizing its current environmental demands.

Keywords

💡Artificial Intelligence (AI)

Artificial Intelligence refers to the simulation of human intelligence in machines that are programmed to think like humans and mimic their actions. In the context of the video, AI is presented as a double-edged sword, with the potential to both contribute to and mitigate the climate crisis through its capabilities in data analysis and decision-making. The script discusses how AI can optimize energy use in buildings and assist in environmental research, yet also consumes significant resources and energy, contributing to environmental challenges.

💡Climate Crisis

The term 'Climate Crisis' encapsulates the urgent situation of global warming and its impacts on the environment, including extreme weather events, melting ice caps, and rising sea levels. The video script emphasizes the role of AI in addressing this crisis, such as mapping Antarctic icebergs and tracking deforestation, while also acknowledging the environmental footprint of AI's operation, particularly in data centers.

💡Data Centers

Data Centers are large facilities that house computer systems and associated components, such as servers, storage systems, and networking equipment. They are crucial for the operation of AI systems. The script highlights the environmental impact of data centers, which consume significant amounts of electricity and water for cooling, leading to potential conflicts with local communities over resource use.

💡Energy Optimization

Energy Optimization involves making decisions and implementing practices to reduce energy waste and improve efficiency. The video discusses how AI can analyze data from large buildings to optimize energy use, thereby reducing waste and conserving resources, which is vital in the context of the climate crisis.

💡Greenhouse Gas Emissions

Greenhouse Gas Emissions are gases that trap heat in the Earth's atmosphere, contributing to global warming. The script points out that the operation of AI, particularly in data centers, can lead to significant greenhouse gas emissions due to the energy consumed, which is often generated by burning fossil fuels.

💡Sustainable Development

Sustainable Development refers to development that meets the needs of the present without compromising the ability of future generations to meet their own needs. The video script touches on the concept of sustainable AI development, questioning whether the current practices of AI, particularly in data centers, align with principles of sustainability and the environment.

💡Biochar

Biochar is a type of charcoal produced from biomass that can be used to enrich soil and sequester carbon dioxide from the atmosphere. The script mentions a biochar facility that Microsoft has engaged with to offset carbon emissions, illustrating one method companies are exploring to mitigate the environmental impact of their operations, including AI and data centers.

💡Carbon Offsets

Carbon Offsets are a form of compensation for the emission of greenhouse gases by investing in projects that reduce emissions elsewhere. The script questions the effectiveness of carbon offsets as a solution to the environmental impact of AI, suggesting that they may not fully negate the emissions produced by data centers and AI operations.

💡Regulatory Bodies

Regulatory Bodies are organizations that establish rules and regulations to ensure compliance with laws and ethical standards. The video script discusses the role of regulatory bodies like the European Parliament in setting AI rules that consider environmental impact, indicating a growing recognition of the need for oversight in this area.

💡Climate Neutral

Climate Neutral refers to achieving a state where the net carbon emissions are zero, often through a combination of reducing emissions and offsetting the remaining emissions. The script mentions initiatives like the Climate Neutral Data Center Pact, which aims to have data centers operate in a climate-neutral manner by 2030, highlighting the industry's efforts to address environmental concerns.

💡Computational Cost

Computational Cost refers to the resources required to perform computations, including energy, time, and hardware. The video script notes the increasing computational cost of training and running AI models, which has implications for the environmental impact of AI as it scales up and becomes more prevalent.

Highlights

AI's potential to help solve the climate crisis is discussed, but it's also acknowledged as part of the environmental problem due to its energy and resource consumption.

AI consumes significant resources like water and electricity, which raises concerns about its environmental impact, especially as technology advances.

Plans for a Google data center were halted due to local concerns over the potential depletion of drinking water resources.

AI's capability to analyze vast amounts of data quickly can optimize energy use in buildings, potentially saving up to 40% of energy and CO2 emissions.

Adian, a German startup, uses AI to help facility managers run commercial buildings more efficiently by analyzing building data and resolving errors quickly.

AI's speed in data analysis is invaluable to researchers and scientists seeking climate solutions, such as mapping Antarctic icebergs and tracking deforestation.

The environmental cost of AI includes the energy and water needed for data centers, which can strain local resources and contribute to climate change.

Training a single AI model can produce more CO2 emissions than a car does in its entire lifetime, including manufacturing and fuel consumption.

Microsoft aims to be carbon negative by 2030 and is investing in carbon dioxide removal credits to offset its emissions, including those from AI operations.

The EU AI Act emphasizes the need for AI systems to be developed and used sustainably, though it lacks specific guidelines on environmental impact.

Communities affected by AI operations, such as data centers, should have a voice in AI governance to ensure sustainable practices.

Governments like Singapore's are aware of the strain data centers can put on power grids and have implemented moratoriums on new constructions.

Tech companies are experimenting with green alternatives for data centers, such as using solar and wind power, but most still rely on fossil fuels.

The EU's AI Act is more focused on the digital environment than the natural one, but the climate impact of AI is becoming increasingly significant.

As AI advances, the need for more data centers and the associated environmental consequences will continue to grow.

The future of AI's role in climate change is uncertain, but its current state requires a substantial amount of natural resources for training and operation.

The discussion calls for more regulations to address the environmental impact of AI and considers the need for community involvement in AI governance.

Transcripts

how can I help save the environment it's

a question someone might ask an

artificial intelligence tool like chat

gbt which has made serious gains in

terms of what it's capable of and though

there's talk that AI might hold the

answers to the climate crisis it's also

part of the problem one that's only

going to get bigger as the technology

advances and its use grows if you train

on twice as much data it takes twice as

much energy running AI consumes real

resources water and electricity and so

do we when we extract solutions from it

that's why plans for a Google data

center were recently put on hold locals

realized it could mean the end of

drinking water for their neighborhood

and so we're asking is AI going to help

solve the climate crisis or make it

worse in this episode of transforming

business there's one thing AI can do

that people can't reading and then

analyzing huge amounts of data quickly

there's literally hundreds of thousands

of data points coming in from a large

building like this one into our system

data that can help us make decisions

like how to optimize the use of energy

in a building so that none of it goes to

waste Henri zorf is a chief technology

officer of Adian a small startup based

in Cologne Germany adan's product helps

facility managers run large commercial

buildings like this one as efficiently

as possible imagine you have a facility

manager in the technical building room

and there's an energy system maybe the

heating or Cooling and it's displaying

an error code error code

15 that person would have first to find

the uh manual for the heating or the

cooling and look up what is error code

15 or that person could just ask our

system hey Adian I'm in the technical

building room uh standing in front of a

whatever fison heating I'm getting

displayed error code 15 how do I resolve

it and then our system knows the entire

history of that building datawise that

combined with the generative AI can

actually like search through the history

and then just tell that person okay for

a code 15 in summer you have to do this

and that and by the way he is also the

manual aian says it can help a building

save up to 40% of energy and CO2

emissions in buildings this large making

a small tweak can have a big impact ad

Defan uses AI to do a lot of different

things to help it structure and analyze

the hundreds of thousands of data points

it gets from a building's technical

system to factoring in variables like

outside climate and building

materials each building is individual

meaning it has uh different tenants it

has different usage behaviors it has

also from how it was built different

thermal behavior um if you have a wooden

building as composed as opposed to a

concrete building that's very that's

very different Adian is capitalizing on

ai's ability to read and analyze data in

a sliver of the amount of time it would

take the world's best researchers to do

the same this speed is what makes AI so

valuable to researchers and scientists

looking for solutions to The Climate

crisis scientists are now using AI to

map Antarctic icebergs 10,000 times

faster than humans and to track

deforestation in real time to better

predict weather patterns and to suggest

more efficient Waste Management Systems

there's no doubt AI has the potential to

do good things for the climate but not

everything about it is a gift to the

environment take this hum for example

which residents in Chandler Arizona here

247 it's the sound of a data center

processing the billions of requests it

gets throughout the day think of AI as

the brain and data centers as the body

that supports the brain to work there

are more than 8,000 data centers in the

world according to the International

Energy agency data center energy

consumption is expected to double in

2026 to what it was in

2022 when I started doing AI research a

decade ago I could run most of the AI

systems I was using on my laptop this is

Jesse Dodge he's an AI research

scientist but today we're using

supercomputers right um some of the

large AI systems that people are

familiar with like the chat Bots or the

image generation systems those run on

really large

supercomputers and consume a potentially

very large amount of electricity these

very large amounts of electricity

produce very large amounts of heat and

also that hum you just heard to keep the

data centers from overheating they must

be cooled down and this is usually done

in one of two ways using air

conditioning or water and lots of it

let's say I engageed in a 15 question

conversation with chat gbt over how I

could be more environmentally conscious

experts calculate I would be consuming

about a half liter of fresh water and

this is where AI can be a little

problematic access to clean water is

competing with um let's say local uses

for it this is what got Google into a

bit of hot water soon after it announced

plans to build a $200 million Data

Center in the workingclass neighborhood

of cios in Chile we all use uh Google

search and other Google tools so

initially the neighbors were quite happy

that Google had chosen this area for

building their their data center this is

Sebastian Leed an AI ethics and Society

lecturer at Kings College London they

saw it as as you know synonymous with

progress development a new poll of

innovation in the area but once they

took took a look at Google's

environmental impact report for the data

center they were startled by what they

learned they found out found out at some

point that this Google data center was

going to use 168 L of water per second

in an area facing drought a drought that

is now in its 15th year and caused

elected officials to ration water in the

capital of Santiago but after Fierce

protests from the community the permit

was put on hold a local environmental

Court told Google it needs to modify how

it plans to cool its servers and

Google's plans for a data center in

Uruguay also face push back when locals

learned how much water it would consume

and water isn't the only natural

resource that AI requires it needs a lot

of electricity too and most of that

electricity still comes from burning

fossil fuels which release the

greenhouse gas emissions causing climate

change training a single AI model

produces is more than five times the

amount of carbon dioxide emissions

generated from a car in its lifetime

that's including the emissions to

manufacture the car and its fuel

consumption once it leaves the factory

it's an astounding amount training an AI

model and then ensuring its continued

existence through large data centers is

a massive drain on natural resources and

also drives up what researchers call

embodied carbon so that's going to be

the amount of um carbon it took to let's

say build the

hardware just starting by mining the

rare earth minerals that goes into the

gpus shipping that across the world to

then uh be manufactured into a GPU and

then shipping that GPU to its final

destination at a data center that does

incur a really large environmental

impact it's this impact that companies

like Microsoft are trying to take into

account as they set climate goals

Microsoft says it's aiming to be carbon

negative by 2030 not just neut

but negative and one way it's hoping to

get there is through Bolivia more than

9,000 km from Microsoft headquarters in

Redmond Washington is a biochar facility

operated by exom mad green it turns

forestry waste into something that's

called biochar which is essentially

charcoal exom mad green will produce the

biochar containing carbon dioxide and

bury it underground where it can enrich

the soil and keep keep CO2 from getting

into the atmosphere Microsoft has bought

32,000 tons of carbon dioxide removal

credits but that's a tiny fraction of

its overall annual emissions we don't

know how much of that will grow with

AI That's one way that Microsoft can

continue to expand its AI operation and

data centers while saying it's still on

path to being carbon negative but is

that effective enough or is it just a

form of corporate greenwashing if we do

something like bubon offsets that

doesn't negate the action that we took

meaning that doesn't undo the carbon

emissions that we've produced these two

things don't cancel each other out as AI

advances governments and Regulatory

bodies are trying their best to keep up

this year new AI rules passed by the

European Parliament will go into effect

impacting businesses like a Defan the EU

AI act does reference the impact of AI

on the environment it asks that AI

systems are developed and used in a

sustainable and environmentally friendly

manner though it doesn't really spell

out what that means Chile's AI laws

which were drafted before the EU rules

but aren't nearly as comprehensive don't

address environmental impact

either I think what's concerning is not

only that this is not being addressed um

enough as it should I think but also

that the the voice of the people

affected by it is not considered right

so so even if you look at research the

Press uh quite often they report on the

environmental impact but they don't

report on how the situation can affect

the uh livelihood or the well-being of

the communities participating somehow

within the value chain of artificial

intelligence if operating a data center

for your AI model is the reason why a

community doesn't have access to

drinking water is that still sustainable

we need those voices to participate as

well in the governance of AI so if they

for example is coming up with you know

new regulation uh it would be great to

be able to hear those communities as

well because those communities they are

not against technology or AI but that

what they will say is that if we want AI

it has to be built in dialogue with

local communities Singapore's government

was aware of the strain data centers

would have on its power grid so it

enacted a 4-year moratorium on the

construction of new ones that ended in

2022 and they've now accepted build

proposals from these companies

Australia's air trunk Chinese data

center operator GDs and Tik Tok owner

bite Dan and equinex and Microsoft based

in the

US Dublin's power operator placed a

moratorium on data centers in 2022 they

say the Irish Capital could face

blackouts if new ones are

built in Europe around a 100 data center

operators signed the climate neutral

data center pact promising to be what

the title implies

a climate neutral data center by 2030

but is that even feasible tech companies

are experimenting with green

Alternatives like this Google facility

in Belgium which partly uses solar

energy and in 2021 Microsoft said it had

opened three data centers in Sweden that

run entirely on wind power but most data

centers still run on fossil fuels and

rely on local power

grids this first draft of the eu's AI

Act does more for the digital

environment than the natural one but as

AI continues to advance its climate

impact is becoming harder and harder to

ignore if you train on twice as much

data it takes twice as much energy and

so that pattern has really um spelled

over the last 10 years this dramatic

increase in computational

cost AI is the new kit on the Block and

it's hard to predict what the future

will look like there's no doubt it can

help us make massive gains when it comes

to slowing down climate change but in

its current state an AI model requires

an enormous amount of Natural Resources

first to train and then to run and not

all data centers are dedicated to AI use

but we will need to build more of them

as AI advances and become standard use

in business research and public

administration should regulations do

more to address the environmental

consequences of

AI absolutely well that's what chat gbt

thinks anyway what about you let us know

in the comments

[Music]