Scan barcode
happymille's review against another edition
challenging
hopeful
informative
inspiring
slow-paced
3.5
jkowski's review against another edition
challenging
hopeful
informative
reflective
medium-paced
3.75
algorithm0392's review
2.0
2.5 stars. Nothing in this book has changed my severe skepticism towards effective altruism as a movement, which is unfortunate, since some of its underlying theories are credible.
The biggest turnoff to me continues to be the lack of intellectual humility among the EA movement (or maybe that's philosophy in general), and this seems to be an undercurrent that I've sensed from others as well.
Relies on a fair amount of hand waving in the end notes in addressing some of the biggest counterarguments, as well as expected value arguments where the values and probabilities extend to infinitely large / small values, which has the effect of breaking logic (at least to me). This last point is finally addressed, in appendix IV: "we should not chase tiny probabilities of enormous amounts of value," though this precise argument drives much of the conclusions presented in the book.
The book is helpful, however, in sharpening my own views and convictions and arguments about what I believe, by presenting in such starkness ideas that I vehemently disagree with.
The biggest turnoff to me continues to be the lack of intellectual humility among the EA movement (or maybe that's philosophy in general), and this seems to be an undercurrent that I've sensed from others as well.
Relies on a fair amount of hand waving in the end notes in addressing some of the biggest counterarguments, as well as expected value arguments where the values and probabilities extend to infinitely large / small values, which has the effect of breaking logic (at least to me). This last point is finally addressed, in appendix IV: "we should not chase tiny probabilities of enormous amounts of value," though this precise argument drives much of the conclusions presented in the book.
The book is helpful, however, in sharpening my own views and convictions and arguments about what I believe, by presenting in such starkness ideas that I vehemently disagree with.
bentrevett's review
2.0
This book is basically an introduction to "Longtermism", a bastardization of Effective Altruism (EA) in which the future matters significantly more than the present.
The book itself is very well written (and very well cited/sourced), but I just couldn't find myself really agreeing with the core principles of Longtermism. Recent events (i.e. the collapse of FTX/SBF) have shown that Longtermism is most probably a giant scam that the current "thought leaders" of EA have either fallen for or are just going along with because it seems to be bringing in more money than actually doing something about the present (e.g. climate change, animal welfare, inequality, etc). I am pretty certain that the EA movement (don't call it a cult!) will distance itself from Longtermism in the near future and that they will de-emphasise the main topics of this book as a sort of "thought exercise" that they actually didn't think too hard about.
The book itself is very well written (and very well cited/sourced), but I just couldn't find myself really agreeing with the core principles of Longtermism. Recent events (i.e. the collapse of FTX/SBF) have shown that Longtermism is most probably a giant scam that the current "thought leaders" of EA have either fallen for or are just going along with because it seems to be bringing in more money than actually doing something about the present (e.g. climate change, animal welfare, inequality, etc). I am pretty certain that the EA movement (don't call it a cult!) will distance itself from Longtermism in the near future and that they will de-emphasise the main topics of this book as a sort of "thought exercise" that they actually didn't think too hard about.
rachelevelyn's review
4.0
I really liked how this book began and enjoyed / found really interesting some parts, particularly the focus on ie targeted donation vs individual actions and the population ethics.
I found the structure a bit confusing and disjointed and got a bit lost in some of the more technical parts.
I found the structure a bit confusing and disjointed and got a bit lost in some of the more technical parts.
ddrake's review
challenging
hopeful
informative
inspiring
reflective
medium-paced
4.0
I really liked this, though I think some of MacAskill's arguments are flawed.
Most notably, his argument that larger human populations are inherently better. His valuation of nonexistent people seems wrong to me. His logic seems to rely on implicit apples-to-oranges comparisons between humans that do or will exist, and those that don't or won't.
I also find the expected value computations and conclusions drawn from them to be problematic, particularly in the extremes when one has a very low probability and a very large value; this shows up not just in longtermism arguments, but in AI risk, pandemic risk, and so on. The philosophy types making these arguments don't seem to appreciate statistics, variation, and so on. They treat the numbers with more precision than they deserve. For example, someone might say there's a 1 in 10 million chance that an engineered pandemic causes, say, 10 trillion dollars of damage to the world economy, conclude that the expected impact of that event is 1 million dollars, and that therefore spending less than one million dollars to prevent that is worthwhile. But they are ignoring the error bars -- that 10 trillion dollars' damage might have two orders of magnitude on either side, so that your expected value likewise has a similar factor-of-10000 range.
When your conclusion relies on a figure with such a large range, your conclusion is correspondingly weaker, in my view, and for many of these problems, the values going into them have enormous variation, so much so that there's really little value in the argument.
That said, I'm sympathetic to the general approach; we just need a little more epistemic humility and a more appreciation of the limitation of our input knowledge.
So, I do like this and am very much on board with the idea of orienting ourselves towards the future, and in taking inspiration from prior social movements that produced profound, good, long-term change for humanity.
Do check out similar books in this area! I particularly like [book:The Good Ancestor: A Radical Prescription for Long-Term Thinking|51107158]; for focusing on your own life, [book:Four Thousand Weeks: Time Management for Mortals|54785515].
For a novel that'll get you thinking about this, try [book:The Actual Star|56304414], and for truly cosmic, mind-expanding thinking, [book:Starmaker|145898848].
I'd also recommend [book:Finite and Infinite Games: A Vision of Life as Play and Possibility|189989]; you could summarize this book in the context of that one by saying: what we owe the future is to play the infinite game.
Most notably, his argument that larger human populations are inherently better. His valuation of nonexistent people seems wrong to me. His logic seems to rely on implicit apples-to-oranges comparisons between humans that do or will exist, and those that don't or won't.
I also find the expected value computations and conclusions drawn from them to be problematic, particularly in the extremes when one has a very low probability and a very large value; this shows up not just in longtermism arguments, but in AI risk, pandemic risk, and so on. The philosophy types making these arguments don't seem to appreciate statistics, variation, and so on. They treat the numbers with more precision than they deserve. For example, someone might say there's a 1 in 10 million chance that an engineered pandemic causes, say, 10 trillion dollars of damage to the world economy, conclude that the expected impact of that event is 1 million dollars, and that therefore spending less than one million dollars to prevent that is worthwhile. But they are ignoring the error bars -- that 10 trillion dollars' damage might have two orders of magnitude on either side, so that your expected value likewise has a similar factor-of-10000 range.
When your conclusion relies on a figure with such a large range, your conclusion is correspondingly weaker, in my view, and for many of these problems, the values going into them have enormous variation, so much so that there's really little value in the argument.
That said, I'm sympathetic to the general approach; we just need a little more epistemic humility and a more appreciation of the limitation of our input knowledge.
So, I do like this and am very much on board with the idea of orienting ourselves towards the future, and in taking inspiration from prior social movements that produced profound, good, long-term change for humanity.
Do check out similar books in this area! I particularly like [book:The Good Ancestor: A Radical Prescription for Long-Term Thinking|51107158]; for focusing on your own life, [book:Four Thousand Weeks: Time Management for Mortals|54785515].
For a novel that'll get you thinking about this, try [book:The Actual Star|56304414], and for truly cosmic, mind-expanding thinking, [book:Starmaker|145898848].
I'd also recommend [book:Finite and Infinite Games: A Vision of Life as Play and Possibility|189989]; you could summarize this book in the context of that one by saying: what we owe the future is to play the infinite game.