Always good to have skeptics to stretch your creativity!
So, the counterfactual as it stands right now is that we're giving somebody additional mental powers through high-speed access to software hooked directly into the brain.
We're not assuming this technology includes advanced AI that does not presently exist. We're sticking for the most part with software that we have now, but we would be safe to give x1000 of existing hardware capabilities.
We could give them an internet connection, but let's say that's cheating. We will not allow them to utilize anybody else's genius or just any available database. For now let's say that they can download large, structured data sets which others have built into their brain-interfaced computers in advance, but they cannot access outside sources in real-time.
OK, so someone like this is going to study for a test. They can study in an ordinary fashion, but we can also build dozens of spreadsheets for them to use during the process.
First of all, any time any question relies on vocabulary, they are going to have that piece in place. They will have a definition of every word or unusual phrase at their immediate beckon call. That solves a lot of reading comprehension problems, but maybe not all of them.
What about those questions where there is a passage to read and, for instance, she has to discern something about the author's intentions?
Here, we get to give whatever kind of custom solver we might choose to provide. For example, we can give her an ability to accumulate a score all of the emotional words in the passage.
It's a standardized test, not a general test of problem-solving ability. Therefore, she gets to include a lot of previous test questions and templates for answers in her data. How much of an advantage will this provide?
She is never going to make an error in arithmetic or algebra, and she will be able to perform these functions very rapidly.
She gets to immediately convert different kinds of units, one to another. She gets to use any formula that can be recorded.
On these tests, there are a rather limited number of kinds of logic problems. She has to somehow recognize which kind she is dealing with in order to answer them correctly, but I think we can build her some kind of classifier in Excel.
The incorrect answers on these tests also fall into some discernable patterns. There is nothing to stop us from including rules of thumb in these spreadsheets, and she can use these rules of thumb to winnow out many of the fakes.
The counterfactual is not complete unless we decide how much time she has to practice for the exam, and how many practice tests she can do in advance. I think that her ability to upload many tests and practice them without using her hands to write is going to give her more benefit from practice than most people get.
We also can presume that she has years of practice using her inbuilt software.
I'll grant you this: The ability to extract critical information from a word problem or a passage, and the ability to translate words into algebraic or logical expressions-these things seem a bit trickier. It's hard to say how much canned computer programs can aid the brain in that realm without actually trying to build them.
Excel permits programming, so the scenario does seem to allow us to give her some Watson-like powers in addition to everything else.
I would really like to know exactly how well a version of Watson can extract meaning from passages of text. My understanding is that is how it answered a lot of the history questions on Jeopardy.
Her special skills may not increase her attention span, though. If she does not care about the test, or if here mind wanders and she starts playing mental video games, then no.
Increased mental capacity is not going to help someone who does not want to use it.
So, the counterfactual as it stands right now is that we're giving somebody additional mental powers through high-speed access to software hooked directly into the brain.
Um, no. That's part of the issue -- we're not giving her access to additional mental powers. We're giving her easy, fast, and convenient access to some information tools. Her mental powers remain the same -- if her working memory is limited, it remains limited. Being able to look up things in a second does not imply a large working memory. If she gets confused with longish logical chain...
This is part of a weekly reading group on Nick Bostrom's book, Superintelligence. For more information about the group, and an index of posts so far see the announcement post. For the schedule of future topics, see MIRI's reading guide.
Welcome. This week we finish chapter 2 with three more routes to superintelligence: enhancement of biological cognition, brain-computer interfaces, and well-organized networks of intelligent agents. This corresponds to the fourth section in the reading guide, Biological Cognition, BCIs, Organizations.
This post summarizes the section, and offers a few relevant notes, and ideas for further investigation. My own thoughts and questions for discussion are in the comments.
There is no need to proceed in order through this post, or to look at everything. Feel free to jump straight to the discussion. Where applicable and I remember, page numbers indicate the rough part of the chapter that is most related (not necessarily that the chapter is being cited for the specific claim).
Reading: “Biological Cognition” and the rest of Chapter 2 (p36-51)
Summary
Biological intelligence
Brain-computer interfaces
Networks and organizations
Summary
The book so far
Here's a recap of what we have seen so far, now at the end of Chapter 2:
Do you disagree with any of these points? Tell us about it in the comments.
Notes
Snake Oil Supplements? is a nice illustration of scientific evidence for different supplements, here filtered for those with purported mental effects, many of which relate to intelligence. I don't know how accurate it is, or where to find a summary of apparent effect sizes rather than evidence, which I think would be more interesting.
Ryan Carey and I talked to Gwern Branwen - an independent researcher with an interest in nootropics - about prospects for substantial intelligence amplification. I was most surprised that Gwern would not be surprised if creatine gave normal people an extra 3 IQ points.
And some more health-specific ones.
People have apparently been getting smarter by about 3 points per decade for much of the twentieth century, though this trend may be ending. Several explanations have been proposed. Namesake James Flynn has a TED talk on the phenomenon. It is strangely hard to find a good summary picture of these changes, but here's a table from Flynn's classic 1978 paper of measured increases at that point:
Here are changes in IQ test scores over time in a set of Polish teenagers, and a set of Norwegian military conscripts respectively:
This study uses 'Genome-wide Complex Trait Analysis' (GCTA) to estimate that about half of variation in fluid intelligence in adults is explained by common genetic variation (childhood intelligence may be less heritable). These studies use genetic data to predict 1% of variation in intelligence. This genome-wide association study (GWAS) allowed prediction of 2% of education and IQ. This study finds several common genetic variants associated with cognitive performance. Stephen Hsu very roughly estimates that you would need a million samples in order to characterize the relationship between intelligence and genetics. According to Robertson et al, even among students in the top 1% of quantitative ability, cognitive performance predicts differences in occupational outcomes later in life. The Social Science Genetics Association Consortium (SSGAC) lead research efforts on genetics of education and intelligence, and are also investigating the genetics of other 'social science traits' such as self-employment, happiness and fertility. Carl Shulman and Nick Bostrom provide some estimates for the feasibility and impact of genetic selection for intelligence, along with a discussion of reproductive technologies that might facilitate more extreme selection. Robert Sparrow writes about 'in vitro eugenics'. Stephen Hsu also had an interesting interview with Luke Muehlhauser about several of these topics, and summarizes research on genetics and intelligence in a Google Tech Talk.
For Parkinson's disease relief, allowing locked in patients to communicate, handwriting, and controlling robot arms.
Big ones I can think of include innovations in using text (writing, printing, digital text editing), communicating better in other ways (faster, further, more reliably), increasing population size (population growth, or connection between disjoint populations), systems for trade (e.g. currency, finance, different kinds of marketplace), innovations in business organization, improvements in governance, and forces leading to reduced conflict.
In-depth investigations
If you are particularly interested in these topics, and want to do further research, these are a few plausible directions, some inspired by Luke Muehlhauser's list, which contains many suggestions related to parts of Superintelligence. These projects could be attempted at various levels of depth.
How to proceed
This has been a collection of notes on the chapter. The most important part of the reading group though is discussion, which is in the comments section. I pose some questions for you there, and I invite you to add your own. Please remember that this group contains a variety of levels of expertise: if a line of discussion seems too basic or too incomprehensible, look around for one that suits you better!
Next week, we will talk about 'forms of superintelligence', in the sense of different dimensions in which general intelligence might be scaled up. To prepare, read Chapter 3, Forms of Superintelligence (p52-61). The discussion will go live at 6pm Pacific time next Monday 13 October. Sign up to be notified here.