Many effective altruists of my acquaintance expect an artificial general intelligence (AGI) to be created within our lifetimes. They expect its abilities to far outstrip our own. Many of them believe there is a high chance of a very positive outcome: the creation of a post-scarcity society where people live for hundreds or thousands of years (if not indefinitely) and no longer suffer from human limitations.

I have occasionally seen people suggest that effective altruists who focus on the far future should not donate kidneys. I think this is a true claim for some sets of beliefs that far-future EAs sometimes have: for example, if you think there is a high chance of human extinction, donating a kidney is less good. But for the beliefs I outlined, I think donating a kidney is a thing more people should consider doing.

Kidney donation increases the patient’s lifespan by about nine years. For people who do not expect a post-scarcity society to be created in the relatively near future, the benefits of donating a kidney are, well, about nine years of life, plus the improved quality of life from not having to be on dialysis.

However, for people who expect a value-aligned superintelligence soon, kidney donation significantly increases the chance that a person will live a very very long lifespan in utopia. (Many people expect that the superintelligence will be able to upload minds or 3D print kidneys or similar.) Rather than a mere 14 QALYs, kidney donation produces hundreds if not thousands of expected QALYs.

Short AI timelines also reduce the costs of donation. Many risks are at the end of life, such as the increased risk of kidney disease; if you have short AI timelines, those risks are irrelevant (either because you’re dead or because the superintelligence has cured kidney disease). While you might have to avoid ibuprofen, you only have to avoid it for ten years or so, at least if you timed your kidney donation correctly.

I am uncertain about when to time your kidney donation relative to your AGI timelines. On one hand, you don’t want to risk overshooting and wasting your kidney; on the other hand, you want to maximize the chance that your recipient will survive until the creation of AGI.

There are certain people for whom it is not the right decision to donate a kidney. An incomplete list:

  • People who are not allowed to donate a kidney by the medical establishment.
  • People who wish to save their kidney to increase the chance that family members or friends make it to the creation of a superintelligence.
  • People who are unusually scared of needles or hospitals or for whom this is otherwise an unreasonably large request.
  • People who are doing direct work, if they expect three weeks of their work to produce more QALYs than donating.
    • It may be worth considering whether the enforced rest from donating a kidney would have some of the benefits of taking a vacation for you.
  • People who do not get paid medical leave or vacation and who either are paid enough that their donations outweigh the benefit of donating a kidney or would suffer financial difficulties from taking three weeks off.

But in general, I think more EAs who are concerned about AGI should consider kidney donation as part of their portfolio of altruistic behavior.