It would be wise for us to employ more people and less artificial intelligence. AI requires many finite critical materials, creates geopolitical and economic wars, and end-of-life hazardous waste. As for people, they use less energy, are abundant, can teach each other, and, at the end of life, they are organic and decompose.
On the
monitor, I received the wording that this reflection stimulates debates about
digital sustainability, where the "logic is clear”, according to the LLM I
used.
"AI is not “immaterial”: it demands enormous
amounts of energy, critical materials (lithium, cobalt, rare earths), cooling
water, and generates technological waste.
Scarcity and geopolitical risks: the demand for
critical minerals is already causing international tension.
Life cycle: servers and chips have a short
lifespan, creating electronic waste that is difficult to recycle."
Regarding «human capital» (instead of «people», like I used in my prompt), the LLM wrote:
"Human capital is renewable: people can learn,
adapt, and collaborate without generating comparable environmental impacts.
Social efficiency: many tasks that are currently automated could be performed by humans with a smaller ecological footprint, if there is investment in training and valuing work."
It then suggested
a comparative table «AI vs Human Capital» regarding energy,
materials, environmental impacts, flexibility, and what it termed hidden costs.
I accepted.
After the table, it left a conclusion:
“This table shows that AI has a significantly greater ecological and material footprint, while human capital is more sustainable and adaptable.” (bold from the LLM tool)
At a time
when billions are being invested in artificial solutions, incurring monumental
costs for the environment, instead of investing in natural and human capital,
it is ironic that this conclusion comes from an AI tool and does not feature
among global political priorities.
Photo by Monica Pinheiro CC BY-NC-SA (CC). October 30, 2025.
