AI Can Estimate Corporate Greenhouse Gas Emissions (venturebeat.com) 22
An anonymous reader shares a report: In 2015, representatives from more than 196 countries met in Le Bourget, France to sign the Paris Agreement. The legally binding treaty limits global warming to a rise of well below 2 degrees Celsius compared to preindustrial levels, preferably capping warming at 1.5 degrees. While the Paris Agreement doesn't spell out how the undersigned are expected to achieve this goal, some countries have pledged to cut their net climate emissions to zero by 2050. For these and other steps to be successful, reliable data is key. While the ability to evaluate companies' carbon footprints will be critical for countries seeking to comply with the measures, only a fraction of companies currently disclose their greenhouse gas emissions. But researchers at Bloomberg Quant Research and Amazon Web Services claim to have successfully trained a machine learning model to estimate the emissions of businesses that don't disclose their emissions.
The researchers say investors could use this model to align their investments with international regulatory measures and achieve net-zero goals. Some regions, including the European Union, require investors to apply a "precautionary principle" that penalizes non-disclosing companies by overestimating their emissions. "Merely 2.27% of companies filing financial statements are disclosing their [greenhouse gas] emissions according to our environmental, social, and governance (ESG) datasets," the coauthors wrote in a paper. "In order to make a meaningful change, we need to measure who is contributing [greenhouse gases] into the atmosphere and monitor their claims to decarbonize."
The researchers say investors could use this model to align their investments with international regulatory measures and achieve net-zero goals. Some regions, including the European Union, require investors to apply a "precautionary principle" that penalizes non-disclosing companies by overestimating their emissions. "Merely 2.27% of companies filing financial statements are disclosing their [greenhouse gas] emissions according to our environmental, social, and governance (ESG) datasets," the coauthors wrote in a paper. "In order to make a meaningful change, we need to measure who is contributing [greenhouse gases] into the atmosphere and monitor their claims to decarbonize."
So what is the most important feature? (Score:2)
The type of product the business is producing, the amount of money the put into renewables? Or the size of the company or the total value of the company?
Another Eye-Catching Headline (Score:1)
Re: (Score:3)
Yes, the unqualified claim "can estimate" reminds one of William Shakespeare's King Henry IV, Part I:
Glendower: I can call the spirits from the vasty deep.
Hotspur: Why, so can I, or so can any man;
But will they come, when you do call for them?
Test versus train problems and over-fitting a model are serious risks with this kind of system.
Re: (Score:2)
Re: (Score:2)
If you use enough parameters, you can overfit anything to arbitrary accuracy. However, this algorithm is still very useful, because it provides delusional regulators with a convenient way to hallucinate the details of their delusions.
What about blowhard politicians? (Score:2)
Don't need AI for that.
Provably false (Score:1)
There is no such thing as AI. So AI estimating anything at all is not possible. Thank me later, I just saved you all a ton of money from investing in this.
Paris treaty is not legally binding (Score:2)
As we've seen, both China and the US stepped out of the treaty as soon as convenient. Also, the treaty doesn't say how because nobody knows how.
The rest have modified the agreement severely since signing because the treaty references a rulebook that is basically at the whim of the negotiators and since the poor countries aren't seeing any money, they're just saying no and the old world countries don't want to pay so they're changing the rules as to when certain targets need met.
Can you sign a legally bindin
And how is this relevant? (Score:1)
The photons coming from the sun couldn't care less whether the CO2 that absorbs them came from a corporate chimney,
a residential home, a car, or, FWIW, an AI.
It's cute to measure emissions, but it is much more precise to measure fossil fuel production. You can safely assume that approximately 100% of production will end up as CO2 within a few years.
Anybody can *estimate* anything (Score:1)
Reminds me of a character claiming to be a wizard in Shakespeare:
GLENDOWER:
I can call spirits from the vasty deep.
HOTSPUR:
Why, so can I, or so can any man,
But will they come when you do call for them?
How does it do it? (Score:2)
Counting the number of rockets the CEO has?
But what should we do about it? (Score:4, Informative)
Another article on the problem but not on the solution.
I see plenty of people demand that we "follow the science" and yet refuse to do so themselves. Science tells us what are the best solutions. The ones with the lowest CO2 emissions, least cost in limited resources, highest return, safest, most abundant, and available today. Those are onshore wind, geothermal, hydro, and nuclear fission. We'll need some natural gas to ease the transition, as an alternative to petroleum and coal. We'll need to develop synthesized hydrocarbon fuels. There's some options on carbon capture to consider. The experts I've seen don't all agree on the best method of carbon capture but they agree it is important. They all agree that we need nuclear fission power. Anyone opposing nuclear fission power today is not serious about global warming.
We can keep talking about how bad this problem is but it's far more productive to discuss how to solve it.
Re: (Score:1, Troll)
Outrage culture demands we find exactly who to blame and exactly what percentage of the blame they deserve long before we talk about real solutions. Most of our energy is going into new ways to play the blame game. Because even in the face of impending doom, it's much more fun to blame somebody than it is to try putting our heads together and coming up with a way out.
We're doomed if we don't change course. Let me fix that sentence: We're doomed.
Re: (Score:2)
Anyone opposing nuclear fission power today is not serious about global warming.
We are having an election here in Canada. Our Green party has a platform of shutting down all fossil fuel infrastructure and banning nuclear power. I would never vote for them in any case, but it should not be too much of a surprise why few people actually take them seriously.
We can keep talking about how bad this problem is but it's far more productive to discuss how to solve it.
Most of the solutions I see are something like buy a BEV, wear a sweater indoors in the winter, don't use AC in the summer, eat less tasty animals, don't travel by airplane or ship, and quit buying stuff. The cilice [wikipedia.org] is optional but
Re: (Score:2)
The cilice is optional but recommended.
Yep, there's a term for that...
https://en.wikipedia.org/wiki/... [wikipedia.org]
This is a philosophy that, at least as I see it, made popular among Democrats by the Carter administration. Carter was nearly singlehandedly responsible for the demise of the civil nuclear power industry in the USA, which had influence around the English speaking world because of how connected the US economy was to other nations. Since Carter there's been an anti-nuclear element in Democrat party policy. They don't like nuclear weapons, nucl
Input / output (Score:1)
So essentially, "AI" can take inputs about a company, perform some calculations, and produce some outputs? Wow, what will they think of next!