47. The Environmental Impact of A.I.

Send us a text

10x the Energy of a Google Search: The Hidden Environmental Impact of AI.
 In this episode of The Circular Future podcast, host Stephanie McLarty, Head of Sustainability at Quantum Lifecycle Partners, dives into the environmental consequences of artificial intelligence. This episode uncovers the hidden costs of AI, including its massive energy, water, and material demands, while also exploring how AI can be part of the solution to sustainability challenges. Through insights and actionable strategies, this episode equips listeners to champion responsible AI use and circular economy practices in their businesses and communities.  

Key Takeaways:

• A single ChatGPT query consumes **10x the electricity** of a Google search, highlighting AI's significant energy demands.  
• Data centers powering AI require vast amounts of raw materials, rare earth metals, and water, with global AI water usage projected to exceed Denmark’s annual consumption by 2027.  
• AI has the potential to aid sustainability efforts by optimizing energy use, tracking emissions, and preventing environmental disasters.  
• Promoting circularity through electronics reuse and recycling helps recover valuable materials and reduces e-waste.  
• Individuals and businesses can mitigate AI’s environmental impact by using it mindfully, advocating for sustainable practices, and partnering with certified recyclers like Quantum Lifecycle Partners.

Thanks for listening!

If you like our podcasts, please leave us a review on Spotify or Apple or wherever you get your podcasts from.
Want to be a guest on The Circular Future podcast? Email Sanjay Trivedi at [e-mail]

• Listen on: https://quantumlifecycle.com/podcast
• Follow us on LinkedIn | Facebook

Episode Transcript

Speaker 1 (00:00):
Artificial intelligence has revolutionized
our world, but what can we doabout the unsettling
environmental impact of it?
Welcome to the Circular Future,your access to thought leaders
and innovations to help you be abusiness sustainability
champion, even if it's not yourcore job.
Up I'm your host, stephanieMcLarty, head of Sustainability

(00:39):
at Quantum Lifecycle Partners,your trusted partner in
electronics circularity.
This podcast is over two yearsold now, and that's roughly the
window in which artificialintelligence, or AI, exploded
into public consciousness.
That's about when I learnedabout ChatGPT and experimented

(01:01):
with it months later with thehelp of a colleague.
It was truly mind-boggling whatthis new platform could do, but
, as I subsequently learned, aihas an enormous environmental
footprint that goes largelyunnoticed.
So let's talk about thebenefits and also the impact of

(01:22):
this technological revolutionFor those of us using AI tools.
I think we've all had a momentwhere we were astounded by the
possibilities of it.
Case in point a couple of yearsago, if I wanted to find themes
in something, I would haveplotted everything down and
moved them around until thetheme groupings made sense.

(01:42):
Similar for making summaries Iwould have mapped out the key
elements, reorganized and lookedfor patterns.
Now we can simply use ChatGPTplug in the data set.
Ask ChatGPT to do something foryou and voila, in a matter of
seconds we have it all done forus, or at least a first draft

(02:06):
anyway.
Chatgpt is now being used on aregular basis around the world.
There are 400 million weeklyactive users, with 10 million
paying subscribers.
Chatgpt accounts for around 60%, that's 6-0% of the market

(02:28):
share of all AI tools.
I mean wow, but it's broaderthan just ChatGPT.
We are unquestionably in the ageof artificial intelligence.
Fun fact we're actually in themidst of a fourth industrial
revolution.
Yes, I said fourth.
Here's a little history lesson.

(02:50):
The first industrial revolutionwas in the late 1700s, with the
steam engine bringing aboutrailroads and industrialization
like factories.
The second industrialrevolution was in the late 1800s
and it was powered byelectricity.
The third was in the 1980s andit was the digital revolution.

(03:13):
Think about how computerschanged our lives and followed
by the internet.
And now here we are.
Everything is being shaped byartificial intelligence.
We certainly see thepossibilities in the electronics
disposal and recycling spacebeing able to better identify

(03:33):
and sort products for betterreuse and recycling outcomes, or
identifying hazards withinproducts, which helps keep
everyone safe, being able to runcertain processes in our
business more efficiently.
And hey, I won't forget thefirst time my quantum colleague
showed me how to use ChatGBT toanalyze our green team survey

(03:55):
results.
The tool took about 200 staffresponses, summarized it and
analyzed elements like thedifferences in responses between
our office staff and ourproduction staff in electronics
recycling, and that had neverbeen done before.
And I just want you to know,there actually are AI podcast

(04:15):
recording tools out there.
But yes, this is still my voiceand our producer still edits
the show for real.
With all the possibility, thereis a huge elephant in the room
the environmental impact of AI.
We may not see it overtly, butit's there.

(04:35):
The reality is AI requiresimmense resources for material
input, energy and even water.
First off, to deliver AI, weneed data centers.
These data centers are run byhigh-performance servers and
other hardware.
That all requires a tremendousamount of commodities, even rare

(04:59):
earth metals, in theirmanufacturing process.
To make a single two kilogramcomputer requires around 800
kilograms of raw materials.
On the energy front, a requestmade through ChatGPT consumes 10
times the electricity of aGoogle search.

(05:21):
Yes, you heard that.
When you ask ChatGPT a question, it uses 10 times the
electricity than if he hadsimply searched by Google.
Then there's water.
The data centers need water,particularly to cool the
electrical components.
Water consumption for AI isalready surpassing that of

(05:43):
entire countries, entirecountries.
Ai's global water consumptionis projected to be four to six
times the annual water usage ofDenmark by 2027.
This is at a time when wealready don't have enough clean
water in many parts of the world.
And to top it off, because theuse of AI is growing

(06:07):
exponentially, so are theimpacts, and the impacts are
often distributed unevenly, withcertain regions carrying a
greater burden than others.
Is it all doom and gloom?
No, on the flip side, ai canalso be a solution and help the

(06:27):
environment in a number of ways.
There are lots of examples ofAI helping to track greenhouse
gas emissions and reduce them,to optimize energy usage, to
anticipate failures that couldlead to major environmental
consequences, to wildlifepreservation and so much more.
There's also work being done ata global level.

(06:50):
More than 190 countries haveadopted recommendations on the
ethical use of AI, which alsocovers environmental aspects.
There is work being done on anindustry level, too.
There are advancements inenergy efficiency and cooling in
data centers.
Some data centers are workingtowards net zero, building huge

(07:11):
solar farms or purchasingrenewable energy credits.
Using AI itself to reduce itsown footprint, such as
redirecting traffic togeographies with less
environmental impact, is anotherkey strategy.
So what can we do on a personalor business level about this?

(07:31):
First of all, let's talk aboutthese impacts and share this
message.
Most people don't know.
Secondly, let's be mindful inhow we use AI.
If something can presumably beanswered by a Google search or
even the good old-fashioned wayof talking to someone, think
about doing that.

(07:52):
We can all keep advocating forresponsible and sustainable use
of AI along with its development.
When there is a groundswell ofvoices, action is so much more
likely to be taken.
Finally, in circularity, weneed to optimize the reuse of
electronics and ensureresponsible recycling at the end
of life, using a certifiedrecycler like Quantum to recover

(08:16):
parts and commodities.
We're all part of the solution.
And remember, if you're lookingfor a leader in electronics
reuse and recycling, we'd loveto chat Head on over to
quantumlifecyclecom and contactus.
This is a Quantum Lifecyclepodcast and the producer is
Sandra Trivedi.
Thank you for being a CircularFuture Champion in your company

(08:38):
and beyond.