[keeda]

The environmental impact discussion is always incomplete because it only looks at AI in isolation. LLMs do exactly nothing by themselves, they are always used to facilitate some human endeavor, so we need to take a holistic view of the system.

I had previously done some napkin math (https://news.ycombinator.com/item?id=46984659) which found LLMs + humans are way more efficient than humans alone, but that analysis is by definition simplistic, and as some point out, does not account for outliers like coding. Fortunately, I just came across this source: https://gist.github.com/mdodkins/9b49624855cc41570c9d1012e0d...

That looks accurate and lines up with my previous numbers. A heavy day of Claude Code usage is equivalent to running 0.7 of a dishwasher cycle. Already that sounds pretty decent, but let's put that into context and refine the simplistic analysis a bit. Discussing footprints in terms of [electricity / water / CO2] tuples, that is approximately:

1.3 kWH / 2.6L / 540g

Compare that to a human's daily footprint (US averages, scaled down to an 8-hour workday; the numbers are actually higher if you look only at office use):

10 kWH / 102L / 4KG

As a % of the human footprint, that is ~ 13% / 2% / 13%.

Right off the bat water usage is negligible. But at a minimum if LLM usage saves a human 13% time at a task, we're at breakeven i.e. the LLM resource consumption is compensated by the savings in human consumption. Anything more than 13% and we actually start conserving resources on average!

Now I know the productivity impact of LLMs is a contentious topic, but my past comments include studies (from 2024, in the era of spicy autocomplete, long before coding agents!) showing ~30% boost in throughput and/or time savings. Interestingly the survey-based https://www.genaiadoptiontracker.com/ finds a consistent 30% time savings across industries. A human completing a task 30% faster consumes ~30% less resources i.e. we save ~3.3 kWH / 34L / 1.3KG per day.

This gives us a net savings of 2 kWH / 31L / 760g per human per day!

Even with 2024 estimates of productivity numbers we are conserving resources! I would wager that a full day's use of Claude Code today offers a much, much higher boost.

You could say coding is a small part of software engineering, but then resource consumption scales down correspondingly. The upshot remains that for a given task overall resource consumption goes down with increased LLM usage! In fact, if LLMs were adopted across all knowledge workers to do all possible tasks, we could recoup the enormous environmental cost of training models in a few business days! Subsequent ongoing savings can more than compensate for a lot of the slop produced elsewhere.

So -- and tongue only partly in cheek -- if you really want to make an environmental impact, use LLMs more.

[thyristan]

> Right off the bat water usage is negligible. But at a minimum if LLM usage saves a human 13% time at a task, we're at breakeven i.e. the LLM resource consumption is compensated by the savings in human consumption. Anything more than 13% and we actually start conserving resources on average!

That ignores the fact that the then-unemployed human will not change it's resource consumption in any way, because a human coding or a human sitting around idly do have roughly the same resource footprint.

Except of course if you are arguing for doing the Skynet thing and terminate all humans replaced by AI...

[keeda]

If you had some goal to achieve -- whatever it was, professional or personal -- would you rather expend X resources to achieve it or (X - Y), where Y was a significant % of X?

And if those resources had an environmental impact, and if you were truly concerned about the environment, why would you want to choose the option that expends more of those resources?

Think of it this way: if you had a doubt you wanted to resolve, would you drive to the library and spend hours or days scanning through reference material... or would you do a Google search?

That is essentially what the "environmental impact" debate comes down to, but nobody is talking about it that way. We can talk all day about the impact of AI on jobs and society, but the impact on the environment is a red herring, and a destructive counter-productive one at that, as TFA shows.

[thyristan]

Yes of course, the capitalist way of thinking forces lower resource consumption, as long as those resources are measured in monetary expenses.

But when thinking about society and the environment, there are different considerations that make the simplistic capitalist thinking wrong in certain situations. Human labor is such a resource that cannot be strictly evaluated by monetary values.

For example (caveat: slight intentional alteration): > would you drive to the library and spend hours or days scanning through reference material... or would you do a [prompt and have AI provide you with a summary]?

Of course on the first glance, as a lazy person myself, the choice is obvious. But is it really? Doing the research yourself has secondary benefits, such as higher familiarity with and learning of the topic. Even as a boss who pays money for people who do that, I'll sometimes gladly pay money so they'll learn and become better employees.

> That is essentially what the "environmental impact" debate comes down to, but nobody is talking about it that way.

Because the environmental impact of human labor behaves differently than the environmental impact of other resources.

If you consume less steel the environmental benefit is clear. Less consumption means less extraction, less processing, less waste and pollution, etc. There is also a direct feedback loop through our capitalist system to induce this. Nobody will mine iron ore that won't be paid for, processed, sold as steel. In that sense, steel is a flexible resource, because there is a direct relation between demand, supply and environmental impact.

If you consume less human labor, things are not so clear and easy. There is a base rate of the "production" of humans, that is relatively independent from the flow of money (actually, it might even be an inverse relation: the less money you have, the more you procreate). The production of human labor is also a slow process, since it takes around 2 decades for a human to grow up. So the resource "human labor" is basically an always-there-resource at some amount that won't be influenced by market forces. The environmental impact of that amount of humans won't be influenced by market forces. So for the environment, it is actually better to give those humans something to do, and use human labor as an always-there-resource, instead of using flexible resources. Same thing for society, humans with nothing to do are generally considered a problem, so you need to give them something to do.

So in a strictly capitalist sense, you are right. But when talking about the environment and society, human labor is a special resource that cannot be treated like other resources.