Resources

Training large language models is an incredibly power-intensive process that has an immense carbon footprint. Keeping data centers running requires a ludicrous amount of electricity that could generate substantial amounts of greenhouse emissions — depending, of course, on the energy’s source. Now, the Verge reports, Microsoft is betting so big on AI that its pushing forward with a plan to power them using nuclear reactors. Yes, you read that right; a recent job listing suggests the company is planning to grow its energy infrastructure with the use of small modular reactors (SMR)…

But before Microsoft can start relying on nuclear power to train its AIs, it’ll have plenty of other hurdles to overcome. For one, it’ll have to source a working SMR design. Then, it’ll have to figure out how to get its hands on a highly enriched uranium fuel that these small reactors typically require, as The Verge points out. Finally, it’ll need to figure out a way to store all of that nuclear waste long term…

Other than nuclear fission, Microsoft is also investing in nuclear fusion, a far more ambitious endeavor, given the many decades of research that have yet to lead to a practical power system. Nevertheless, the company signed a power purchase agreement with Helion, a fusion startup founded by OpenAI CEO Sam Altman earlier this year, with the hopes of buying electricity from it as soon as 2028.

In 2007, the U.S. Congress mandated the blending of biofuels such as corn-based ethanol into gasoline. One of the top goals: reducing greenhouse gas emissions. But today, the nation’s ethanol plants produce more than double the climate-damaging pollution, per gallon of fuel production capacity, than the nation’s oil refineries, according to a Reuters analysis of federal data. The average ethanol plant chuffed out 1,187 metric tons of carbon emissions per million gallons of fuel capacity in 2020, the latest year data is available. The average oil refinery, by contrast, produced 533 metric tons of carbon.

The ethanol plants’ high emissions result in part from a history of industry-friendly federal regulation that has allowed almost all processors to sidestep the key environmental requirement of the 2007 law, the Renewable Fuel Standard (RFS), according to academics who have studied ethanol pollution and regulatory documents examined by Reuters. The rule requires individual ethanol processors to demonstrate that their fuels result in lower carbon emissions than gasoline. The Environmental Protection Agency (EPA) is charged with writing the regulations to meet the goals set by Congress. For processors, that translates to an EPA requirement that the plants use certain emissions-control processes the agency assumes will result in lower-than-gasoline emissions. But the agency has exempted more than 95% of U.S. ethanol plants from the requirement through a grandfathering provision that excused plants built or under construction before the legislation passed. Today, these plants produce more than 80% of the nation’s ethanol, according to the EPA.

Some of the exempted plants produced much less pollution, including some owned by the same companies producing the highest emissions. The EPA said about a third meet the law’s environmental standard even though they are not required to do so. But as a group, the plants freed from regulation produced 40% more pollution per gallon of fuel capacity, on average, than the plants required to comply, the Reuters analysis found.

China successfully powered up its “artificial sun” nuclear fusion reactor for the first time, state media reported Friday, marking a great advance in the country’s nuclear power research capabilities. Phys.Org reports:
The HL-2M Tokamak reactor is China’s largest and most advanced nuclear fusion experimental research device, and scientists hope that the device can potentially unlock a powerful clean energy source. It uses a powerful magnetic field to fuse hot plasma and can reach temperatures of over 150 million degrees Celsius, according to the People’s Daily — approximately ten times hotter than the core of the sun. Located in southwestern Sichuan province and completed late last year, the reactor is often called an “artificial sun” on account of the enormous heat and power it produces. They plan to use the device in collaboration with scientists working on the International Thermonuclear Experimental Reactor — the world’s largest nuclear fusion research project based in France, which is expected to be completed in 2025.

The amount of energy required to “mine” one dollar’s worth of bitcoin is more than twice that required to mine the same value of copper, gold or platinum, according to a new paper, suggesting that the virtual work that underpins bitcoin, ethereum and similar projects is more similar to real mining than anyone intended.

One dollar’s worth of bitcoin takes about 17 megajoules of energy to mine, according to researchers from the Oak Ridge Institute in Cincinnati, Ohio, compared with four, five and seven megajoules for copper, gold and platinum.

Other cryptocurrencies also fair poorly in comparison, the researchers write in the journal Nature Sustainability, ascribing a cost-per-dollar of 7MJ for ethereum and 14MJ for the privacy focused cryptocurrency monero. But all the cryptocurrencies examined come off well compared with aluminium, which takes an astonishing 122MJ to mine one dollar’s worth of ore.

It is well-established established that Bitcoin mining — aka, donating one’s computing power to keep a cryptocurrency network up and running in exchange for a chance to win some free crypto — uses a lot of electricity. Companies involved in large-scale mining operations know that this is a problem, and they’ve tried to employ various solutions for making the process more energy efficient.

But, according to testimony provided by Princeton computer scientist Arvind Narayanan to the Senate Committee on Energy and Natural Resources, no matter what you do to make cryptocurrency mining harware greener, it’s a drop in the bucket compared to the overall network’s flabbergasting energy consumption. Instead, Narayanan told the committee, the only thing that really determines how much energy Bitcoin uses is its price. “If the price of a cryptocurrency goes up, more energy will be used in mining it; if it goes down, less energy will be used,” he told the committee. “Little else matters. In particular, the increasing energy efficiency of mining hardware has essentially no impact on energy consumption.”

In his testimony, Narayanan estimates that Bitcoin mining now uses about five gigawatts of electricity per day (in May, estimates of Bitcoin power consumption were about half of that). He adds that when you’ve got a computer racing with all its might to earn a free Bitcoin, it’s going to be running hot as hell, which means you’re probably using even more electricity to keep the computer cool so it doesn’t die and/or burn down your entire mining center, which probably makes the overall cost associated with mining even higher.