03/06/2026
What does an aligned artificial general intelligence (AGI) mean for animals? Animals are often overlooked, even among those who take transformative AI seriously.
That’s why the EA Forum spent a week discussing the question earlier this year. Below, we share three picks from the week: considering factory farming, wild animals, and concrete ways we can make a difference for animals.
1️⃣ "AI Safety and Cross-Species Robustness: A brief critical review"
Jim Buhler argues that reducing existential risk from AI isn't necessarily good for animals: saving humans means the preservation of animal farming, and the case for outweighing this reality elsewhere is weaker than it looks. https://forum.effectivealtruism.org/posts/dpMSoisajgK22peRt/ai-safety-and-cross-species-robustness-a-brief-critical
2️⃣ "If AGI goes well for humans, will it go well for (wild) animals?"
Most animals live in the wild, so it’s worth considering whether a human-aligned AGI will improve their lives. Simon Eckerström Liedholm puts the probability at around 30%, warning that current alignment work could bake in indifference to wild animal welfare. https://forum.effectivealtruism.org/posts/kqy256tmnYjnDHwDc/if-agi-goes-well-for-humans-will-it-go-well-for-wild-animals
3️⃣ "List of ideas for improving animal welfare in light of transformative AI"
Michael Dickens lays out 11 concrete ways animal advocates can make a difference for animals in light of transformative AI — placing advocacy to pause AI at the top. https://forum.effectivealtruism.org/posts/d3gaMea82DWCd6wwz/list-of-ideas-for-improving-animal-welfare-in-light-of
Read all 18 posts from the week: https://forum.effectivealtruism.org/topics/agi-and-animals-debate-week
