This is a 104 megawatt bitcoin mining data center
. Inside are 32,000 machines, performing over one
million trillion calculations per second. This facility's 250,000 square feet. A big facility, a lot of machines. But the sound you hear is going to be the fans
themselves that are running. Validating transactions on the bitcoin network and
mining new coins essentially involves guessing a winning number, which at this moment takes on
average over 100 sextillion tries. All that work helps to secure the network by
making it nearly impossible for bad actors to accrue enough computing power to take control. But in 2020, mining bitcoin also consumed 75.4
terawatt hours of electricity, more than all of Austria or Portugal. Sometimes it's been as much as Norway, which is
about 120 terawatt hours, so it's a lot of electricity. Recently, the world's second largest
cryptocurrency, Ethereum, moved away from the massive computational guessing game known as
proof-of-work to a new method of validating transactions. Ethereum has migrated to a system known as
proof-of-stake, which swaps out miners for validators. So instead of running these large
banks of computers, validators leverage their existing cache of ether as a means to verify
transactions and mint new tokens. This move, called the merge, eliminated nearly all
of Ethereum's energy consumption and former Ethereum miners have been left searching for
alternate coins to mine. So currently it's not profitable. So I have my rigs shut off. A lot of miners are selling their cards. While it's technically possible for bitcoin to
move to proof-of-stake too, there are no indications that it will as crypto enthusiasts
disagree on what method is most secure. Instead, some large bitcoin miners are touting
their use of renewables and even see bitcoin's energy use as an asset. We'll be fully carbon neutral at the end of this
year in our operations. I want to invest in more clean power. You want bitcoin mining to keep going in the
United States because every day when we buy power off the grid, that's more money into the utility
grid. But Jones notes that after China banned bitcoin
mining last year, the industry has only gotten dirtier. There may be individual efforts here and there,
but globally the renewable mix, again, has seemed to be going down over time, not up. Bitcoin is the world's first cryptocurrency. And when it was invented in 2009, people mined
coins from their personal computers using less than a second's worth of electricity. Since then, energy use has skyrocketed, but the
purpose of mining has remained the same. When a miner correctly guesses an impossible to
predict number, this serves to validate a group of bitcoin transactions called the block and makes
it impossible to tamper with. As a reward for helping secure the network the
winning miner is given bitcoin. And that is what we know as mining. We see millions of devices all around the world
constantly, by a process of trial and error, try to guess a certain winning number and if they
guess correctly, then they get to create the next block for the blockchain and then they get that
reward for it. But the network is designed such that the more
miners are vying for the prize, the more difficult it becomes to win. On average, one block is mined every 10 minutes,
no matter the price and how many miners are on the network. So as prices soared and more people
joined, miners started using specialized devices called ASICs, which can make over 100 trillion
guesses per second. In May 2021, it took about 13 years worth of an
average household's electricity to mine one bitcoin. And while prices have tumbled since,
mining difficulty and hence electricity usage has actually increased all year, reaching a record
high in October as large scale Ethereum miners switched to mining bitcoin. Today, most mining takes place in industrial
scale data centers like this one in Marble, North Carolina. It's operated by Core Scientific, one
of the world's largest bitcoin miners. Each individual machine today is about 3200 watts. So if you imagine that's like two hairdryers,
probably for one machine here in Marble where we're at today, it's 104 megawatt facility. Marathon Digital Holdings is another large miner
currently building out facilities in West Texas and North Dakota. Today we have deployed an operational upwards of
30,000 machines and will scale that number significantly. So by the middle of next year will
be approaching 200,000 mining rigs installed for a total capacity of about 20 exahash. 23 exahash means 23 quintillion guesses per
second. And while that's a lot, it's only about 9% of the
entire network's current computing power. If any one entity controlled 51% or more of the
network, it could act maliciously and alter the blockchain. But that would likely cost billions. And the very fact that you have to spend resources
in this process also makes it expensive for any attacker that wants to attack this network
because they have to pay for those resources as well. From a security perspective, the system has been
working. Many bitcoin enthusiasts fear alternate methods
would be less secure and want to stick with this tried and true approach. From an energy perspective, though, proof-of-work
has been a lightning rod for critique. But an alternate validation method called
proof-of-stake has been in development for a decade now. Many smaller cryptocurrencies already
operate using proof-of-stake. And since its inception, Ethereum
has planned to migrate to the system too. So Vitalik Buterin, who's one of the co-founders
of Ethereum, has been talking about making the move to proof-of-stake consensus mechanism since
2014, and the transition has been repeatedly pushed back for the last several years for a
variety of reasons, most notably the fact that the tech just wasn't ready. But in September of 2022, after years of
preparation, the network finally made the highly anticipated switch, instantly reducing Ethereum's
energy use by 99.9%. We can see the merge is successful when we receive
the first finalized block, and it happened very quickly. And with the finalization that was done, it did
not show any error at the time. So we quickly realized that the transition has
been smooth. And Ethereum's proof-of-stake system verifying
transactions does not involve a massive competitive guessing game which eats up tons of
energy. Instead, miners are replaced by validators who
are pseudo randomly assigned to propose and verify new blocks of transactions which earns them
rewards. To ensure that these validators act honestly,
they essentially have to make a security deposit staking a certain amount of coins into the
network. Ranjan says this will make the network more
secure than proof-of-work because bad actors will be punished and lose their money. I think the penalties involved are very, very
heavy here, so I don't think it's in their favor to do any kind of notorious activity on the
network. The more coins someone stakes, the better their
chances of being assigned to create and validate blocks. Today, the minimum required stake is 32
ether, which for most of this year has been worth over $40,000. However, it is possible to stake much less by
participating in staking pools. Since the merge, some miners like Aaron Petzold,
who has accumulated about $25,000 worth of mining equipment over the years, have been looking to
repurpose their graphics cards to mine altcoins instead. When the merge hit, I immediately switched my
everything to ergo, in which I was making, I think, profiting maybe $10 a day, which is
nothing compared to when Ethereum was going strong and very quickly within about two days so much
hash rate diluted the market of all these different coins. Raven coin, flux, ergo, Ethereum classic to name
a few where it's no longer profitable. Well, Petzold used to mine bitcoin. It's no longer profitable for him to do that
either. And while he has some Ethereum staked right now,
he's holding out on staking more while he evaluates the market. Though Ethereum's price has actually fallen since
the merge, Petzold is still bullish on its long term prospects. I do believe we will see a $4,500 Ethereum one day
and a $10,000 Ethereum one day. I don't know if it'll be in three weeks, three
months, three years or 30 years, but I believe in the technology and I'm optimistic about the
future. Yet he does have some qualms about proof-of-stake. Overall, my opinion is I like I like proof-of-work
because it's worked since the beginning. Why mess with something that's been working and
doing very well? One issue with proof-of-stake, I feel, is it
leads to centralization. Concentrating power in the hands of those with 32
or more Ether to stake is a common concern, though others say the risk of centralization is no
different than with Bitcoin mining, where those with the most computing power are most likely to
win rewards. Either way, the rich get richer, and if either
the largest staking pools or mining pools wanted to join forces, they could theoretically wreak
havoc on their respective networks. But while questions of security and
centralization are up for debate. There's no doubt that proof-of-stake is a massive
improvement from an energy efficiency standpoint, essentially eliminating Ethereum's carbon
footprint, while Bitcoin still uses as much energy as some small countries. Benjamin Jones, associate professor of economics
at the University of New Mexico, says he's often asked how damaging bitcoin is compared to other
commodities like gold or crude oil or solar power. So he authored a paper on it. People say bitcoin, digital gold, let's just say
it's digital gold. What are the climate damages of gold? How do they compare to bitcoin? And we also did
this for like crude oil, beef, chicken, pork, natural gas, renewables, wind, solar, precious
metals like platinum and copper. Jones and his colleagues found that from 2016 to
2021, each $1 in Bitcoin market value created was responsible for an average of $0.35 in global
climate damages due to the emissions associated with mining. For comparison, a dollar's worth of
gold produces only $0.04 of climate damages, while natural gas is responsible for $0.46. Crude oil burned as gasoline is responsible for
$0.41 and beef is responsible for $0.33. Over the years, Jones studied the climate damages
of bitcoin actually exceeded the coin's value about 6% of the time. We're able to estimate this because we have a
pretty good sense of where miners are located when they sign up for mining pools. And so we get information on their IP addresses
and then we can apply country specific emissions factors for carbon and other pollutants to give
us some sort of daily measure of the carbon emissions every day to mine bitcoin around the
globe. China, which previously hosted up to 75% of global
bitcoin mining capacity, kicked miners out in mid 2021 as the country saw crypto as a threat to
both its central banking system and to a lesser extent, its environmental goals. This led to dirtier mining overall, though,
because while China has an abundance of hydropower, many miners moved to Kazakhstan,
reminding is largely coal powered, and the U.S., which is now the world's largest bitcoin miner. Nobody seems to expect bitcoin to move to
proof-of-stake any time soon, though such a massive change would require massive buy in from
the network and many bitcoin enthusiasts and investors see proof-of-work as a better proven,
more secure system. They think that proof-of-work mining is the only
way to make a secure cryptocurrency, so they are currently not making any moves towards
implementing a proof-of-stake based system. Large miners who've made the upfront investment in
mining equipment also have a financial stake in the proof-of-work model. But now some are trying to retool and rebrand as
renewable energy champions. As we move to become fully carbon neutral, by the
end of this year, our energy mix will be a mix primarily of wind energy with some grid energy,
and that grid energy will offset the fossil fuel portion of that with renewable energy credits. Previously, Marathon was notorious for reviving an
old coal plant in Montana to power its bitcoin mining operations. Now the company has closed its Montana plant and
is partnering with a wind farm in Texas, purchasing excess wind energy that would
otherwise go to waste. This is a wind farm where we operate behind the
meter, so we're not sitting on the grid taking energy directly off the grid. We're behind the meter at the source of energy
generation at this wind farm and only use grid energy when the wind is blowing, if you would. Global estimates for the amount of carbon free
electricity used in mining vary widely, though one industry group reports that nearly 60% is carbon
free. While Cambridge University, which publishes the
daily Bitcoin Electricity Consumption Index, found that in January of this year, about 38% came from
renewables or nuclear power. That's about the same percent as the U.S.'s
overall electricity mix. And so I think nailing that number down is really
important. But the problem is how do you get the data right? Because this is anonymous at the end of the day. And so it's really hard. It's a really challenging to do that. Jones's Research shows that given his data from
2016 to 2021, if Bitcoin mining used 93% carbon free electricity, then the climate damages as a
share of coin price would drop to 4%. Putting bitcoin's damages on par with solar power
or gold. But no matter whose renewable energy estimates,
we believe that's a long way off from where we stand today. Core Scientific's data center in
Marvel runs fully on grid electricity. Executive VP of Client Services, Russell Cann,
says that's actually beneficial because when there's high electricity demand, such as during a
heat wave, bitcoin miners can power down their operations in an instant. So, you know, power company says, listen, we have
this excess energy and 8300 hours out of the year, we're going to sell it to you at a very
inexpensive rate because these other 300 hours a year we're going to turn you off because we need
that generation in order to go and feed the rest of the grid and feed to hospitals and schools. And while, say, not, mining bitcoin at all would
also lessen strain on the grid, Cann says that ultimately investing in the grid through these
large data centers will allow utilities to build more energy infrastructure, ideally green
infrastructure. That's money to say, look, we need another 100
megawatts and we want it to be solar, or we need more nuke, or we need wind and solar combo or
need more hydropower. So if I want to have green power, I'm going to
invest in more bitcoin mining and I'm going to encourage it to come to the United States versus
somewhere else. While Ethereum's price has dropped after the
merge, its founder, Vitalik Buterin, speculates this could just be an example of the phrase "buy
the rumor, sell the news," where hype around an event drives up the price of an asset only to
have it drop later on. And while bitcoin's price remains low, more
miners have flocked to the network in the wake of the merge. As data centers, space and electricity
previously devoted to Ethereum mining has been freed up. That means bitcoin is consuming more
electricity than ever before. But so long as the network continues to operate
smoothly and securely, the move will likely serve as a signal to regulators, crypto enthusiasts and
skeptics alike that the industry can transition to sustainability. So slashing energy consumption by more than 99%
will also go a long way toward lowering the barrier to entry for institutional investors who
have been battling the optics of contributing to the climate crisis. The White House released a report warning that
proof-of-work mining operations could get in the way of efforts to mitigate climate change so it
definitely helps with optics. The recent report from the Office of Science and
Technology Policy analyzed the climate and energy implications of crypto assets in the U.S. and put fourth a series of recommendations that
include developing environmental performance and energy efficiency standards for cryptocurrencies,
tracking and analyzing the energy usage and fuel mix of crypto miners and greater transparency and
reporting these metrics. I think what just happened with Ethereum switching
from energy intensive proof-of-work to negligible energy use, proof-of-stake was a game
changer. And I think that's where, if regulation is going
to have an impact, it's how do we get bitcoin to make that switch? New York State even passed a bill in June that
would place a moratorium on crypto mining in the state if it's not powered by renewables. But the governor has yet to sign the bill into
law. Meanwhile, some former Ethereum miners like
Petzold are still hoping they'll be able to find another use for their mining equipment. I'm optimistic that there'll be another coin that
that's GPU mineable because there's just so much equipment out there right now that there's
someone that I'm sure that wants to take advantage of all this equipment they can use to secure
network. That's just a no brainer. In all likelihood, energy intensive mining for
bitcoin, in particular, is likely to continue for the foreseeable future. So now Jones says it's just a matter of figuring
out what exactly the damages are and what damages society is willing to accept when it comes to an
emerging technology like crypto whose full value to society is yet unknown. A lot of things use electricity and a lot of
things have impacts on the climate, but I think it's just putting it into context what the
damages are. And bitcoin is particularly a damaging, relatively
speaking, commodity.
