3 ms·

Let's say there's a 1% chance that Neuralink is able to improve the quality of life for paralyzed people by 100% (double) within 20 years, at the cost of 10,000

Let's say there's a 1% chance that Neuralink is able to improve the quality of life for paralyzed people by 100% (double) within 20 years, at the cost of 10,000 high-order animals (sheep, pigs, monkeys, etc). Is that worth it? If they could do it with only 1000 animals instead, but this delayed availability by 10 years, would that be better? These are the types of questions we need to think about in order to have a productive conversation. Black-and-white positions with binary and objective moral superiority are not possible. We must weigh costs and benefits, and acknowledge that we might end up in different places based on having different, but reasonable and good-faith priorities.

danShumway 4y ago
1. It doesn't sound like the extra animal suffering is actually helping their odds, it sounds like if anything it's going to reduce customer trust in Musk's ability to implant electronics in their brain without killing them. 2. It's possible to do this research without this extreme level of reckless testing on live animals. 3. It's not clear that Neuralink's approach will actually be better for society in the long-term. Musk's policies towards consumer freedom and privacy in every single one of his businesses he's been a part of from Paypal to Tesla has been atrocious; I would prefer another company that's not owned by Musk build neural implants because I don't trust Musk not to screw the people who are getting the implants. The correct questions here are: - Will sacrificing 1,500 animals actually increase Neuralink's odds of doubling the quality of life for paralyzed people in 20 years by 1%? I don't think it will. - Will other companies be able to produce the same results in 20 years without that suffering? Probably, yes. Nueralink is not the only company researching this. - Do we want Neuralink to "win" the market of neural implants in the first place? I would argue that we probably don't, the best-case scenario for neural implants would be if they came out of an Open ecosystem in cooperation with more traditional companies who have less of a startup/move-fast-and-break-things vibe. - Is the sacrifice that Neuralink wants to make generalizable as a moral principle? No, we know what it looks like when animal testing is deregulated, and it looks awful, and Musk shouldn't be given a special pass on those rules. The problem with doing quick math on whether suffering is "worth it" is that it can be used to shut conversations about how to get the same outcomes with less suffering. It's an argument that's been in play basically ever since activism around animal testing began, and repeatedly, the people who argue "it's worth it to save everyone" have been proven wrong -- both because it's been repeatedly proven that we can do the same testing without Musk's extreme level of disregard for animal life, and because the outcomes very often turn out to be over-exaggerated or gloss over serious caveats. People are very bad at estimating the odds of success of a venture, and very bad at estimating what the potential benefits or downsides of an outcome will be, and so animal testing without moral boundaries tends to lead towards people killing a bunch of animals for no good reason. In practice, saying "anything goes as long as the research is important enough" hasn't really born out well morally, so it's good for us to have standards that apply to everyone, regardless of what they are working on or how important they think the project is.
aradox66 4y ago
That kind of reasoning is not really relevant, as it can justify all kinds of unnecessary suffering. A more relevant question is whether neuralink is following best practices in setting up these studies and carefully justifying the use of each animal. Animal studies are already widely accepted as a justifiable necessity in medical trials. That doesn't imply a blanket endorsement of well intentioned but sloppy use of animals.
SketchySeaBeast 4y ago
I feel like you could do that exact same math with human lives as well, but either way the reasoning is flawed - there is no benefit to sloppy mistakes that produce unusable data.

danShumway 4y ago

1. It doesn't sound like the extra animal suffering is actually helping their odds, it sounds like if anything it's going to reduce customer trust in Musk's ability to implant electronics in their brain without killing them. 2. It's possible to do this research without this extreme level of reckless testing on live animals. 3. It's not clear that Neuralink's approach will actually be better for society in the long-term. Musk's policies towards consumer freedom and privacy in every single one of his businesses he's been a part of from Paypal to Tesla has been atrocious; I would prefer another company that's not owned by Musk build neural implants because I don't trust Musk not to screw the people who are getting the implants. The correct questions here are: - Will sacrificing 1,500 animals actually increase Neuralink's odds of doubling the quality of life for paralyzed people in 20 years by 1%? I don't think it will. - Will other companies be able to produce the same results in 20 years without that suffering? Probably, yes. Nueralink is not the only company researching this. - Do we want Neuralink to "win" the market of neural implants in the first place? I would argue that we probably don't, the best-case scenario for neural implants would be if they came out of an Open ecosystem in cooperation with more traditional companies who have less of a startup/move-fast-and-break-things vibe. - Is the sacrifice that Neuralink wants to make generalizable as a moral principle? No, we know what it looks like when animal testing is deregulated, and it looks awful, and Musk shouldn't be given a special pass on those rules. The problem with doing quick math on whether suffering is "worth it" is that it can be used to shut conversations about how to get the same outcomes with less suffering. It's an argument that's been in play basically ever since activism around animal testing began, and repeatedly, the people who argue "it's worth it to save everyone" have been proven wrong -- both because it's been repeatedly proven that we can do the same testing without Musk's extreme level of disregard for animal life, and because the outcomes very often turn out to be over-exaggerated or gloss over serious caveats. People are very bad at estimating the odds of success of a venture, and very bad at estimating what the potential benefits or downsides of an outcome will be, and so animal testing without moral boundaries tends to lead towards people killing a bunch of animals for no good reason. In practice, saying "anything goes as long as the research is important enough" hasn't really born out well morally, so it's good for us to have standards that apply to everyone, regardless of what they are working on or how important they think the project is.

aradox66 4y ago

That kind of reasoning is not really relevant, as it can justify all kinds of unnecessary suffering. A more relevant question is whether neuralink is following best practices in setting up these studies and carefully justifying the use of each animal. Animal studies are already widely accepted as a justifiable necessity in medical trials. That doesn't imply a blanket endorsement of well intentioned but sloppy use of animals.

SketchySeaBeast 4y ago

I feel like you could do that exact same math with human lives as well, but either way the reasoning is flawed - there is no benefit to sloppy mistakes that produce unusable data.