Hacker News new | past | comments | ask | show | jobs | submit
Here's a handy calculator you can use to estimate how much CO2 and water I wasted with my coding agent session: https://www.andymasley.com/visuals/ai-prompt-footprint/
Not sure what point you wanted to make, but this calculator is quite shocking. GPT 5.5 pro, with "a long document" and 10 requests a day gives 25% of daily CO2 emissions!

Ten coding sessions a day with Opus is still 4.7%!

This feels enormous. I will definitely stop rolling my eyes when people complain about AI CO/water usage...

GPT-5.5 Pro is a notoriously expensive model, it's 6x the price of GPT-5.5. Not something to use as a daily driver!

That ten coding sessions a day with Opus number feels more credible to me.

What are you on about? May be 1 out of 100,000 users are using 5.5 Pro to make 10 "Long Documents" as defined in that tool EVERY day. What a silly thing to harp on.

Six 100,000 token Claude coding sessions use less energy than a dryer load, and less water than making one egg. If you are truly concerned about energy and water usage, AI is not even in the top 100 things you should be concerned about in your daily life.

This very obtusely ommits the demand for new data centers and related infrastructure that using AI creates, the going "vegan for a year" option assumes less cows being born but somehow the "don't use AI" doesn't assume that the data center wasn't build in the first place.
The discrete number of cows being born is theoretically fine-grained enough to actually respond to 2–3 vegans yielding one fewer cow. It's unlikely on a one-year time scale, but one cow only goes so far.

Even a thousand AI objectors aren't going to limit the demand for a data center, in no small part because these investments are only partially driven by current demand and are significantly driven by expectation of future demand. And they're really not going to lead to smaller data centers either because if you're building a data center in the first place you're going to spec it out for future demand.

Regardless, I think in both cases it's important to be realistic about the actual impact that one person has. If that number is disappointingly small, that serves as signal that your conscientious objection isn't making the industry you're objecting to as uncomfortable as you would like to think. It may still be worth objecting for your own sense of self, or maybe it serves as an invitation to evangelize your position more, but either way there's not much value to measuring things in a way that gives you an illusion of greater impact than you actually have.

The real point is not "one session", it's the fact that people now do that routinely, that CICD are using those to check every commit, and each search engine query now does that too, so it multiplies