Genetic Optimization Like picking cards for your poker or bridge hand

Bill Hsu is a scientific advisor to BGI (formerly, Beijing Genomics Institute) and a founder of its Cognitive Genomics Lab. He is an expert on the genetics of intelligence and what might be possible for enhancing intelligence via genetics. His insight also represents what is possible for genetically enhancing other characteristics.

There are positive correlations between narrow abilities. This suggests that an individual who is above average in one area (for example, mathematical ability) is more likely to be above average in another (verbal ability). They also suggest a robust and useful method for compressing information concerning cognitive abilities.

Normal random genetic can mean that someone can still be very, very good at physics but poor at verbal communication or at least not a genius at communication. However, there were still people like Carl Sagan who was a genius at science and an excellent communicator.

Arguments against the probable effectiveness of genetic cognitive enhancement are like the arguments against radical life extension, they are based upon non-enhanced examples of today. If you get to pick the cards for your poker or bridge handsthen clearly the bad options can be avoided. The games are different than being dealt random cards.

Bill Hsu believe super-intelligence via genetic enhancement might be a distant prospect, but smaller, still-profound developments are likely in the immediate future. Large data sets of human genomes and their corresponding phenotypes (which are the physical and mental characteristics of the individual) will lead to significant progress in our ability to understand the genetic code—in particular, to predict cognitive ability. Detailed calculations suggest that millions of phenotype-genotype pairs will be required to tease out the genetic architecture, using advanced statistical algorithms. However, given the rapidly falling cost of genotyping, this is likely to happen in the next 10 years or so. If existing heritability estimates are any guide, the accuracy of genomic-based prediction of intelligence could be better than about half a population standard deviation (meaning better than plus or minus 10 IQ points).

Each one-point increase in a country’s average IQ, the per capita GDP was $229 higher. It made an even bigger difference if the smartest 5 percent of the population got smarter; for every additional IQ point in that group, a country’s per capita GDP was $468 higher. A population wide boost of 10 IQ points would be a 10-15% economic boost to countries like the USA.

There are some results which suggest that the genetic basis of intelligence involves 20 times more genes than the genetic basis of height. This means that more genomes will need to be studied to properly genetically decode intelligence. The new CRISPR gene modification method seems up to the challenge of genome engineering and being able to make thousands of genetic modifications. Also, the means of delivering into the body are being vastly enhanced with progress towards surgical nanorobots.

Arxiv – On the genetic architecture of intelligence and other
quantitative traits [30 pages]

I gave a talk at Transhuman Visions a couple of weeks ago.

In my look at near term transhumanism, I see older Tiger Moms as being the driver of early adoption of genetic intelligence enhancement and the lifting of the One child policy in China.

China’s One child policy is being lifted just as embryo selection based upon intelligence for invitro fertilized (IVF) babies becomes possible and we are on the cusp of genetic engineering. Women in China who are now older were banned from having babies but now will be allowed to have children. Many will not be able to conceive naturally and will use IVF. I see IVF going from 400,000 per year worldwide to 2-8 million per year over the next 10 years. IVF babies are more easily embryo selected and accessible for genetic modification. This would provide an economic boost to China in 20-30 years and the beginnings of a significant societal shift.

* Older women use IVF more than younger women.
* Societal shifts that cause more older women to use IVF to have children means more opportunity for embryo selection and genetic intelligence enhancement.
* Countries that permit embyro selection and genetic intelligence enhancement provide the opportunity for IVF to be used for enhancement
* Medical tourism to permissive countries is another means for older women to use IVF in combination with embryo selection or genetic enhancement.

In my talk I first summarized enhancement of human capabilities via products that we can buy. (Smartphones, Google Glass, Apple Siri, IBM Watson, forklifts, cars etc…) As those things get better, anyone can adopt them by buying the latest versions.

Steroids do enhance performance. They work. About 10 million people use them and it is primarily because of vanity. However, we do not live in the wild west or have to compete as Gladiators in Rome. The more powerful people in the world are the billionaires. In the real world the equivalent of Montgomery Burns from the Simpsons has more power than the equivalent of Captain America. Burns can hire his own police force or mercenaries. It is his lawyers and lobbyists who do his work.

100 years ago it was vaccines that began altering the physical attributes of people in a meaningful way. It extended lifespans and improved health. Health improvement boosted productivity and GDP.

Studies show that more intelligence leads to better lives for people. They go to jail less. Their jobs are better and they divorce less.

Each one-point increase in a country’s average IQ, the per capita GDP was $229 higher. It made an even bigger difference if the smartest 5 percent of the population got smarter; for every additional IQ point in that group, a country’s per capita GDP was $468 higher. This is according to Are the Wealthiest Countries the Smartest Countries? (Heiner Rindermann, of the Chemnitz University of Technology, Psychological Science )

However, we can see the damage when intelligence is lower across large national populations.

48% of children in India are stunted. Diseases can leave brain damage when they do not kill. This reduces IQ points by 11-20 on average across the country. This makes India more poor.

Embryo Selection

* In Vitro Fertilization (IVF) currently is used for 400,000 babies each year.
* A new “Alka Seltzer method” for controlling PH reduces costs from $10K-20K in developed world to about $1K
* Developing world was $2-4K and can be $200
* Comprehensive Chromosome Screening (about $6000) can boost successful IVF by 10-20%

–> Lower costs for IVF means more adoption and widespread basic genetic screening via chromosome screening will already be used for high IVF success rate

* China lifting one child ban
* This will mean annual birthrates will go from about 16 million ==> 25.5 million
* 23-42 normal child bearing age
* IVF at older ages up to 50-55 now at about 5-15% but improving
* More older women in China will use IVF
* China is not culturally against enhancing children (a lot of tiger moms and cultural differences)
* Lower cost and more effective IVF with standard chromosome screening, easy step to gene screening
* Maybe 4 million IVF/year with gene screening

Children tend to fall within a spread of 13 IQ points above and below the average IQ of their parents.
Positive outlier at around 2 or 3 percent where child is two standard deviations above parents
Pick the smartest genome from a batch of, say, 20 embryos (but it could 50 or more embryos) to get 20-30 IQ points higher
We are technologically close to non-destructive sequencing of human gametes and zygotes by sequencing 10-20 cells.

Genetic Engineering for Intelligence

BGI (Beijing Genomics institute has a large intelligence study of thousands of geniuses
Various studies finding genes with up to 0.5% impact on intelligence
Intelligence is 40-80% inheritable
There are likely hundreds to a thousand genes that genetically determine intelligence
Similar to height. According to Steve Hsu’s estimates (based on actual data) most humans have (order of magnitude) 1000 rare (-) alleles for intelligence and height

One standard deviation above average has (very roughly) 30 fewer (-) variants.
No negative alleles might be 30 SD above average! Such a person has yet to exist in human history…
Each standard deviation (SD) up or down are defined as 15 IQ points greater or less,
95 percent of the population scores an IQ between 70 and 130, which is within two standard deviations of the mean.
30 SD above average would be and IQ of 550.

Maybe IQ 1000 is Impossible or Meaningless

550 IQ would be like a 13 foot tall person
Physiology limits practical height
What are intelligence limits ?
Brain structure
Average human height is 70 inches and 3 inch SD (standard deviation)
8 feet 1 inch – 97 inches this is 9 Standard Deviations over average
235 IQ is the equivalent in intelligence of a 8 foot 1 inch person

Geniuses and Society

5% of population with 30 points higher intelligence might be about $14000 more GDP per capita
5% of population with 120 points higher intelligence might be about $56000 more GDP per capita
What would a society with tens of millions of Edisons, Einsteins, Steve Jobs and Elon Musks be like ?
Could we get beyond them in capability?

5% of population significantly intelligence enhanced would be possible if IVF takes off and embryo selection and genetic engineering with it over the next 10-20 years.