wdmacaskill.bsky.social @wdmacaskill.bsky.social · 25/02/2026I’m excited to say that the revised 10-yr anniversary edition of Doing Good Better is out now! It’s got updated statistics and a new foreword, reflecting on the last ten years and responding to some key criticisms. 190
wdmacaskill.bsky.social @wdmacaskill.bsky.social · 08/01/2026Almost no one has articulated a positive vision for what comes after superintelligence. What should we be trying to aim for? Utopias from history look clearly dystopian to us, and we should expect the same for our own attempts. We don’t know enough, or have the authority, to decide the details. 211
wdmacaskill.bsky.social @wdmacaskill.bsky.social · 08/08/2025Sometimes people argue that material abundance will make people more altruistic: they’ll get all their self-interested needs met, and the only preferences left to satisfy will be altruistic. 100
wdmacaskill.bsky.social @wdmacaskill.bsky.social · 08/08/2025Should we expect widespread moral progress in the future? In a new paper, Convergence and Compromise, @FinMoorhouse and I discuss this. Thread. 151
wdmacaskill.bsky.social @wdmacaskill.bsky.social · 06/08/2025How easy is it to get to a future that’s pretty close to ideal? In “No Easy Eutopia”, @finmoorhouse and I argue that it’s very hard - very unlikely to happen by default, without some beings in the future trying hard to figure out what “goodness” consists in, and actively trying to promote it. 150
wdmacaskill.bsky.social @wdmacaskill.bsky.social · 04/08/2025Today I’m releasing an essay series called Better Futures. It’s been something like eight years in the making, so I’m pretty happy it’s finally out! It asks: when looking to the future, should we focus on surviving, or on flourishing? 271
wdmacaskill.bsky.social @wdmacaskill.bsky.social · 24/03/2025The claim isn't at all that a century of tech progress in a decade would make the world worse. It's that it would create distinctive challenges that we should prepare for. We even say this, clearly! 120
wdmacaskill.bsky.social @wdmacaskill.bsky.social · 17/03/2025These have big implications for power distribution: a software intelligence explosion would concentrate power (likely in the US); a full-stack one would spread it out. 111
wdmacaskill.bsky.social @wdmacaskill.bsky.social · 17/03/2025A crucial insight: even if a software intelligence explosion doesn't happen or fizzles quickly, there could still be AI-technology and full-stack intelligence explosions. These would start more gradually, but might become very fast. 121
wdmacaskill.bsky.social @wdmacaskill.bsky.social · 17/03/2025Our analysis suggests each feedback loop could by itself lead to accelerating AI progress. And before hitting physical limits, effective compute could increase by 24 orders of magnitude or more - enabling huge improvements in AI capabilities. 111
wdmacaskill.bsky.social @wdmacaskill.bsky.social · 17/03/2025We identify three key feedback loops in AI development: - Software: AI develops better algorithms + data - Chip technology: AI designs better computer chips - Chip production: AI and robots build more computer chips 210
wdmacaskill.bsky.social @wdmacaskill.bsky.social · 13/03/2025How will the creation of AGI be governed? CERN and the Manhattan project are often suggested as models for an international AGI project. We think an under-discussed international organisation, Intelsat, provides a promising complement to these existing models. See 🧵1/ 140
