• lime!@feddit.nu
    link
    fedilink
    arrow-up
    1
    ·
    edit-2
    1 day ago

    i’ve provided mistral with sources and asked it to summarize them, then do averages on those summarisations. i’ve not just asked it to pull data from wherever (except the scrapping thing). i too think ai’s are full of shit, but i can go back and check because i provided the data.

    the assumptions made in your text give a gCO2/kWh figure of about half the IEA’s, where’s that from? also, the emissions numbers of your fossil fuels engines are way off. assuming a fuel consumption of 10l/100km, the number for a petrol car would be 230g CO2/km rather than 150. also you’re mixing your units a lot.

    • Log in | Sign up@lemmy.world
      link
      fedilink
      arrow-up
      1
      ·
      edit-2
      22 hours ago

      Yeah, like I say, AI is shit. Quoting it as an authoritative source is crazy, and AI is surprisingly bad at arithmetic.

      The graphic is screwy. It doesn’t pass plausibility test. Somehow the carbon cost of generating green electricity is far higher than the carbon cost of extracting oil and refining it. Someone’s adding in a whole bunch of CO2 for manufacturing and installing some wind turbines but not for oil extraction machinery and oil refineries. Just the sheer quantity of steel alone isn’t even comparable.

      So no, you can chatgpt your argument as much as you like but you can’t convince me that the cheapest greenest wind overnight electricity I power my EV with somehow took more CO2 by quite a margin than the oil extraction, oil tankers and oil refining that my neighbour’s diesel car does. That’s so backwards and obviously incorrect and I don’t know why you persist with entertaining the idea.