Ray writings | book foreword: Virtually Human

March 1, 2016

Dear readers,

I recommend this article in USA Today profiling Martine Rothblatt, PhD’s keynote at South by Southwest.

USA Today | Sirius founder envisions world of cyber clones, tech med

Her talk featured a roundup of concepts about the future of the human brain and the potential for people to interact through virtual avatars and recreations of an individual.

USA Today said about her keynote AI, immortality and the future of selves:

“Visionary pharma tycoon and futurist Martine Rothblatt, PhD told a packed audience during her South by Southwest keynote speech that eventually these advances in software will rise to the level of consciousness.

“Rothblatt is founder of Sirius Satellite Radio, chief executive of United Therapeutics, and a transhumanist philosopher who believes technology will one day grant humans eternal life.

“Rothblatt’s keynote described how emergence of cyber consciousness, when machines act with a sophistication and thought level equal to that of humans, will not be overnight but subtle evolution.

“With gains in software and robotics, cyber technology will push the envelope of human existence, she said, with the chance to keep living well past traditional limits.”

Martine Rothblatt and I share many ideas and have collaborated on singularity related projects.

I recently contributed the foreword to her book Virtually Human, seen below.

Ray Kurzweil

on the web | essentials

Wikipedia | Martine Rothblatt, PhD

on the web | background

South by Southwest | main
South by Southwest | YouTube channel

South by Southwest | Artificial intelligence, immortality and the future of selves — keynote by
South by Southwest | Beyond the cloud: big data in the off planet era — panel w. Martine Rothblatt PhD

book | foreword

book: Virtually Human
author: by Martine Rothblatt PhD
foreword: by Ray Kurzweil

Virtually Human: a foreword to the book by Ray Kurzweil

In her book Virtually Human, Martine Rothblatt, PhD provides a compelling and convincing case for virtual humans. After all, what difference does it make if our mental circuits are biological or electronic if the result is the same?

She stakes out the scientific case that we will see such humans within a small number of decades and persuasively examines the philosophical and social implications. Both she and I have been articulating this case since we met fifteen years ago.

In my 1999 book The Age of Spiritual Machines (ASM), I made the scientific case that we will see human-level intelligence in a machine by 2029.

These artificial intelligences (AIs) will be capable of passing the Turing test, Alan Turing’s eponymous exam to determine if an AI is indistinguishable from a biological human (to biological human judges) using an instant messaging conversation.

A conference of AI experts was held at Stanford shortly after the publication of ASM and the consensus then was that human-level AI would indeed happen but not for hundreds of years.

Several lines of criticism of ASM emerged such as “Moore’s law will come to an end,” “hardware may indeed be expanding exponentially but software is stuck in the mud,” “consciousness and free will are impossible in machines,” “human level AI may be feasible but is not desirable for biological humans,” and others.

I wrote The Singularity is Near (TSIN) to address these criticisms, which was published in 2005.  In 2006, a conference called “AI 50,” was held at Dartmouth to mark the fiftieth anniversary of the 1956 Dartmouth conference that gave artificial intelligence its name, and the consensus at that meeting was that human-level AI was only 25 to 50 years away.   I’ve stuck with my 2029 prediction, which is now a median view and there is a growing group of people who think I am too conservative.

One piece of evidence of the expanding power of AI is IBM’s Watson, which won a televised Jeopardy! Contest against Brad Rutter and Ken Jennings, the best two (biological) human players in the world.  Indeed Watson got a higher score than Rutter and Jennings combined.  Critics often like to dismiss the significance of AI by saying that it may be good at narrow skills such playing chess or driving a car but machine intelligence does not have the broad and subtle powers of biological human intelligence.  But Jeopardy! is not so narrow a task.  It involves the ability to reason over all human knowledge and the queries are presented in natural language including puns, metaphors, riddles, and jokes.  For example, Watson got this query correct in the rhyme category: “A long tiresome speech delivered by a frothy pie topping.”  The query stumped Rutter and Jennings, but Watson quickly responded: “What is a meringue harangue?”

What is not widely appreciated is that Watson’s knowledge was not hand coded by the engineers.  It got its knowledge by reading Wikipedia and several other encyclopedias, all natural language documents.  It does not actually read these documents as well as you or I.  It might read one page and conclude that “there is a 56 percent chance that Barack Obama is President of the United States.

You might read that page, and if you didn’t happen to know this ahead of time, conclude that there is 98 percent chance.  So you did a better job at reading and understanding that page.  But Watson makes up for its relatively weak reading by reading more pages, a lot more, 200 million pages in all.  And it has a good Bayesian reasoning system to combine all of its inferences so it can conclude overall that there is a 99.9 percent like hood that Obama is President.  And it can do this type of reasoning on all 200 million pages that it has read in the three second Jeopardy! time limit.

Thus one significance of AIs actually reading at human levels, which I maintain will happen by 2029, is that they will then be able to combine their human-level understanding with Internet scale and thereby apply that comprehension to tens of billions of documents.

So what will the significance be of the advent of human-level AI?  A lot of science futurism movies such as Terminator conclude that these AIs will have little use for biological humans.  But if we examine the trajectory of AI, indeed the entire history of invention, we can come to a different conclusion.  Thousands of years ago, we were unable to reach the fruit at that higher branch, so we fashioned a tool that extended our physical reach.  We then created tools that expanded the strength of our muscles so we were able to build giant pyramids in the desert.

Today, we can access all of human knowledge with a few keystrokes with devices we hold close to our bodies.  And the distribution of contemporary AI is not limited to a few wealthy corporations or government agencies but are in billions of hands.  We have thus expanded our physical and mental reach, and that will continue to be the case as AI at human levels and beyond become a reality.

A key message of ASM and TSIN is that the price-performance and capacity of information technologies expands at an exponential pace, currently doubling about every year, a phenomenon that I call the “law of accelerating returns.”  At the same time, the physical size of these technologies is shrinking at a pace of about one hundred in three-dimensional volume each decade.  So computational devices in the 2030s will be the size of blood cells and we will introduce them into our bodies and brains noninvasively.

One application will be to health.  Artificial T-cells will expand the capability of our immune system.  Today our biological immune system does not recognize cancer (it thinks that it’s part of you) and is unable to cope with retroviruses.  We will be able to finish the job with a non biological immune system that will download new software from the Internet to deal with new pathogens.

These “nanobots” will also go into the brain via the capillaries and connect our neocortex (the outer layer of the brain where we do our thinking) to the cloud.  So today, just as we can access many thousands of computers in the cloud when we need them, in the 2030s and beyond we will be able to access additional neocortex to think deeper thoughts.

In my recent book, How to Create a Mind, I describe the neocortex as a self-organizing system of about 300 million modules, each of which can learn, remember and process a pattern.  These modules are organized in a hierarchy and we create that hierarchy with our own thinking.  Only mammals have a neocortex, so when the “Cretaceous Extinction Event” (a violent sudden change in the worldwide climate probably caused by a meteor) occurred 65 million years ago, the ability of the neocortex to quickly devise and master new skills resulted in mammals overtaking their ecological niche.

Another significant event occurred 2 million years ago: the evolution of humanoids with a large forehead, which allowed for a significant expansion of the neocortex.  This additional quantity of pattern recognition modules was the enabling factor for our species to invent language, art, music, science and technology.

We are now on the verge of expanding our neocortex again.  The panoply of devices we carry with us is already expanding the power of our brains.  Indeed I felt that a  part of my brain had gone on strike during that one day “SOPA” strike (when services such as Wikipedia and Google went on strike to express opposition to new privacy legislation).  In the 2030s we will directly expand the size and scope of our neocortex into the cloud.  The only difference this time is that the expansion will not be limited to a certain physical size, but will continue to expand exponentially.  And remember what happened the last time we expanded our neocortex when we became humanoids two millions years ago.  That quantitative expansion enabled a profound qualitative leap and this will happen again.

Rothblatt’s Bina48 is an outstanding example of recreating the physical and mental reality of an actual human in a machine.  Having met the biological Bina Rothblatt, her robotic avatar is not yet equivalent, but is wonderfully suggestive of what is to come.

In my books, I make the case that recreating the computational capacity of the human brain requires about 10^14 (10 to the fourteenth power or 100 trillion) calculations per second.  We already have that capacity in our supercomputers and personal computers will have that power in the early 2020s.  The software for human-level intelligence will take longer but we are also making exponential gains in modeling and recreating the powers of the neocortex.  Creating synthetic models of the neocortex is what I am currently working on as a Director of Engineering at Google.  I make the case in my books that we will have the software capabilities for human-level AI by 2029.  Watson is already a significant milestone in that effort.

Once that is possible, we will be able to create specific personalities including those of people who have passed.  Rothblatt’s Terasem Foundation is devoted specifically to this scenario, a prospect that is thoroughly examined in this book. The movie The Singularity is Near (TSIN), which Martine Rothblatt was executive producer of (and which I wrote) examined this idea, as did Transcendent Man, a movie about my ideas by filmmaker Barry Ptolemy.  That movie illustrates my efforts to preserve the documents, music and other memorabilia of my father so that future AIs can create an avatar with his memories, skills and personality.  Spike Jonze based his recent movie Her on my books and the movies TSIN and TM.

The heroine of Her is an AI (which in the movie is called an Operating System or OS) named Samantha whose voice is provided by Scarlett Johansson.  Even though Samantha is non biological, she is sufficiently human to be able to fall in love with Theodore, the biological protagonist and for Theodore to fall in love with her.  The movie also borrows Rothblatt’s and my idea of creating an avatar to bring back biological humans who have passed in the form of Alan Watts, the poet and philosopher from the 1960s.

Ultimately we will be able to access the information in our brains that constitutes our memories, skills and personalities and back them up.  In my timeline, that is a 2040s scenario. One way that will happen is that by the mid 2030s our thinking will be a hybrid of biological and nonbiological thinking.  The nonbiological part (largely in the cloud) will be subject to my law of accelerating returns. Thus by the 2040s, the nonbiological part of our thinking will greatly predominate.  It will be capable of fully understanding and modeling the biological part.  And it will be fully backed up just as we back up all nonbiological processes today.

Human-level AI is close at hand.  The prospects that Rothblatt writes eloquently about in this book may seem daunting today but so did the idea of a massive network of communication that would tie together virtually all humans when I wrote about that prospect in the 1980s.  When these new technologies do occur, it is remarkable how quickly we accept them as part of everyday reality and cannot imagine how we ever lived without them.

related reading:
Dartmouth College | Dartmouth Artificial Intelligence Conference: The Next Fifty Years — AI at 50