Computing power per dollar has doubled every ~2 years for the past 40 years, per Moore's Law.
Can biology keep up? The next generation of humans would reach adulthood in 20 years, at which time computers would have ~1024x today's processing power. That's a pretty high bar for selective breeding or genetic modification to keep up.
Humans share our values and can generally be overwhelmed with sheer numbers should they not. Making them is unlikely to be dangerous. The same cannot be said for unfriendly AIs. We still have no idea how to make a friendly AI, and it could easily be a century before we begin to have an idea. Even if biology cannot keep up, improving intelligence in the short run will have positive effects on human productivity in the short run, which will get the goal of making a friendly AI closer.
I'm not saying that biological modification will ultimately bring about singularity or even extremely dramatic changes in human capabilities. Rather I think it will address many talent shortages simultaneously, for not that much money. I'm proposing it as an idea we can implement now.
AI research seems to be intelligence constrained. Very few of those capable of making a contribution are aware of the problem, or find it interesting. The Berkeley-MIRI seminar has increased the pool of those aware of the problem, but the total number of AI safety researchers remain small.
The goal of MIRI isn't to increase the speed of AI research but to increase FAI research. Speeding up AI research on the whole would likely increase Xrisk instead of decreasing it.
However, the impact is likely to be limited due to the cost of these methods which will prevent them from having population-wide influence
Why do you think so?
Look at the cost of IVF: According to http://www.momjunction.com/articles/much-ivf-treatment-cost-india_0074672/ it is $450,000 Rs, which is $6,000 dollars. IVF is a prerequisite for the sort of genetic tampering we are talking about, unless you want to use rabies as a carrier. IVF is widely practiced and has few barriers to entry, making me think it won't get much cheaper. That is a lot of money for many parts of the world. To think this cost will come down in the next 10-20 years significantly requires believing that significant advances in automation of the process are possible: that might be true.
Financial incentives to have smarter children are likely to work better in those regions where $6,000 is a lot of money. It's possible that combining both strategies works even better.
Over the years I've read several fascinating books on the nature of intelligence in both human and nonhuman animals; sadly, I don't have the titles at my fingertips. There is no agreement among experts on how to define intelligence, and it is widely recognized that standard IQ and aptitude tests do not encompass all aspects of the topic. It seems pointless to me to expend much effort in increasing human intelligence until the topic is better defined. I think that providing people with analytical thinking skills and the encouragement to use them is likely to deliver better outcomes for humanity.
Thanks to the efforts of organizations like MIRI, four years ago I began adapting work I performed in the value of information to AI safety. I am much indebted to these organizations for increasing public awareness of the topic. I describe this adaptation process here: http://tinyurl.com/h4ttwuo
There is no agreement among experts on how to define intelligence,
Actually IQ has broad support from a lot of experts.
and it is widely recognized that standard IQ and aptitude tests do not encompass all aspects of the topic.
IQ doesn't need to encompass all aspects of the topic to be a quite useful metric.
Do you oppose the EPA decision to reduce mercury pollution because it lowers children's IQ on the grounds that IQ isn't a good measurement of intelligence? They should rather not increase the average IQ of the population through regulation?
Do you oppose the EPA decision to reduce mercury pollution because it lowers children's IQ on the grounds that IQ isn't a good measurement of intelligence?
The possible effect of environmental pollutants on human health (mental and/or physical) is another fascinating and extremely complex topic. I'll avoid venturing into these deep waters on this particular thread.
The possible effect of environmental pollutants on human health (mental and/or physical) is another fascinating and extremely complex topic.
It's not another topic. It the same topic of expanding effort making decisions to increase IQ. Mercury poluttion doesn't kill or decrease lifespan significantly but it reducdes IQ. If you don't accept the existance of IQ as a valid measurement the EPA case for regulating mercury falls flat.
I'll see if I can find the books I referred earlier regarding intelligence testing for people interested in delving further into this topic.
EDIT: One interesting factoid I recall - IQ tests were originally developed to detect impaired mental function only. However, performance on these tests is now used to justify claims of superior mental function. As I recall, among experts this use of IQ testing is controversial.
IQ testing is controversial in some ways but supported in others.
In support of IQ, some forms of IQ tests ('g' loaded tests) tend to reproduce similar scores for the same individual. Further, this score is linked to various life outcomes - higher numbers of patents created, higher academic success rates, higher income, less time in jail, etc. As well as all of this, IQ has been found to be hereditary through twin studies. Lots of literature on this suggest that whatever IQ measures, even if it's not intelligence, it's useful to have in western societies.
But here's why it's controversial: Firstly, there is a potential gender and racial bias. Certain races tend to do better than others on average even controlling for socioeconomic status and the like. Men tend to be at the extreme ends of the scale, with many more falling into the high scoring ranges (2+ standard deviations) than women as well as in the low scoring ranges. Secondly, langauge barriers are another large problem with any verbal-based IQ test, which restrict those tests ability to accurately gauge a test taker who is writing with English as their non-native tongue. Thirdly, there are arguments about how a single number could accurately represent all of human intelligence. In tangent with this, there is debate about what constitutes intelligence, and how we should group it. Should emotional intelligence count? Should physical (kinetic) intelligence count? Should math count as much as verbal? Should problem solving count? Etc.
Against the last point, ignoring the less traditional sorts of intelligence (e.g. kinetic [bodily movement] intelligence), 'g' loaded tests support the idea that even if you're bad at math, if you've got a high 'g' score you'll likely be above average at math if you're high above average in 'g' loaded verbal or logical reasoning tests. So it seems that even if you're deficient or exceptionally good at one area, there is some sort of underlying factor that does help explain at least some difference in the traditional realms. And that underlying factor is what 'g' loaded tests are supposed to assess.
Also worth noting is the decreasing returns after about the second deviation. Although it does continue to have increasing effects in some areas, benefits in other areas start to drop off. It has been argued that IQ can help find a limiting factor but after that limiting factor (around 120-130 IQ on a 15 SD scale) it stops being as useful for prediction. To explain in a better way, "genius" has stopped being linked to a specific IQ. Instead, it's thought that a minimum IQ of around 130 ~ 120 is needed to be a genius, but there is no set point of IQ where you are automatically a genius. You could have 180 IQ and not be a genius, or you could have 120 IQ and be a genius.
As far as I know (I might be wrong) IQ is especially useful at finding exceptionally low-skilled individuals.
So largely it's controversial in that it represents a universal intelligence, It's less controversial that it's some sort of useful construct which predicts a great deal of life outcomes in western society with decent accuracy within the groups it was designed to test (Western-cultured English speakers in particular)
As I recall, among experts this use of IQ testing is controversial.
That depends a lot on who you call expert. There are people with an ideological agenda against IQ and you may have read books arguing against IQ. That doesn't mean that's the psychometic community doesn't still consider IQ test to be valuable. It doesn't mean that IQ isn't a tool valued by organisation like the EPA.
The usage of the EPA is clearly not standard usage of what the test was desgined to do. If you want to argue that you oppose the usages of the test for purposes besides what the original purpose of the tests you attack the basis on which the EPA decision against mercury pollution rests.
I'm writing this article to explain some of the facts that have convinced me that increasing average human intelligence through traditional breeding and genetic manipulation is likelier to reduce existential risks in the short and medium term then studying AI risks, while providing all kinds of side benefits.
Intelligence is useful to achieve goals, including avoiding existential risks. Higher intelligence is associated with many diverse life outcomes improving, from health to wealth. Intelligence may have synergistic effects on economic growth, where average levels of intelligence matter more for wealth then individual levels. Intelligence is a polygenetic trait with strong heritability. Sexual selection in the Netherlands has resulted in extreme increases in average height over the past century: sexual selection for intelligence might do the same. People already select partners for intelligence, and egg donors are advertised by SAT score.
AI research seems to be intelligence constrained. Very few of those capable of making a contribution are aware of the problem, or find it interesting. The Berkeley-MIRI seminar has increased the pool of those aware of the problem, but the total number of AI safety researchers remain small. So far very foundational problems remain to be solved. This is likely to take a very long time: it is not unusual for mathematical fields to take centuries to develop. Furthermore, we can work on both strategies at once and observe spillover from one into the other, as the larger intelligence baseline translates into an increase on the right tail of the distribution.
How could we accomplish this? One idea, invented by Robert Heinlein, as far as I know, is to subsidize marriages between people of higher than usual intelligence and their having children. This idea has the benefit of being entirely non-coercive. It is however unclear how much these subsidies would need to be to influence behavior, and given the strong returns to intelligence in life outcomes, unclear that they can further influence behavior.
Another idea would be to conduct genetic studies to find genes which influence genetics, and conduct genome modification. This plan suffers from illegality, lack of knowledge of genetic factors of intelligence, and absence of effective means for genome editing (they tried CRISPR on human embryos: more work is needed). However, the result of this work can be sold for money, thus opening the possibility of using VC money to develop it. Illegality can be dealt with by influencing jurisdictions. However, the impact is likely to be limited due to the cost of these methods which will prevent them from having population-wide influence, instead becoming yet another advantage the affluent attempt to purchase. These techniques are likely to have vastly wider application, and so will be commercially developed anyway.
In conclusion genetic modification of humans to increase intelligence is practical in the near terms, and it may be worth diverting some effort to investigating it further.