Skip to main content

tv   Book Discussion on Machines of Loving Grace  CSPAN  December 30, 2015 10:33pm-11:31pm EST

10:33 pm
the question is if you have enough to do something with but where would you put them, their habitat was gone and part of that is can you engineer change. >> so this is a good question, the polar bear. you've asked the right person because this is a major theme of research in my lab. polar bears are special case. polar bears have almost no genetic diversity but there is a genome sequence that is 100,000 years old. that polar bear also has almost no genetic diversity of their lack of diversity is not because of any recent decline or recent
10:34 pm
climate change, they are a top predator that is very precisely adapted to their particular habitat. you're you're right, when that habitat disappears, they don't have any diversity which would allow them to adapt to different habitats that are around. the bears have an interesting strategy dealing with this. we discovered recently. not only that it's happening right now for climate change it every time the climate changes rapidly in the past, the same things happen. they breed with brown bears. they produce offspring that go live as brown bears. so the jeans and polar bears but polar bears as this discreet thing. this white bear that is precisely adapted to the arctic, it would disappear is engineering traits that would allow them to survive in other environments is not the case because they'll read with brazen live in that of environment but
10:35 pm
we still not have polar bears. but absolutely. the idea for other species might be you have some that are devoid of diversity in some way, can we increase the amount of diversity in them a genetic shot if you will. that is where this technology is the most powerful. it's what i was talking about in the black footed. ferret
10:36 pm
10:37 pm
were really at the early stages of all of this. it's a technical challenge standing in the way of actually applying with real question. nonetheless, i think it does have the potential to be useful. >> i want to ask about honeybees and the recent issues they been having in their population. >> it would be great if we could find genes that made that resistance, find some bats that are resistant on them figure out what part of a genome and make
10:38 pm
that -- transplant that. that be fantastic in exactly the kind of good application of this technology. i'm not saying we have this and ought to do this, i'm just highlighting. i think that's a great one, areas where we should push this along because it would be incredibly useful. [inaudible] >> the question was how do we date the bones that we have given it stuff that is washed up with sediments. sometimes we get lucky and we find them in the stratigraphic context for their still frozen. in that case we get even luckier and sitting in between all canada cash layers that have been dated. often, we have to generate dates on every single bone.
10:39 pm
we do that with carbon. it's a typical approach. that means were limited by the resolution, the rated carbon which was in the last 50 years. we know we can get dna from stuff older than that. the oldest oldest genome is from a horse that we found in the klondike, in this location of north. it was found with volcanic ash layer that we know is 700,000 years old. the only reason it has any dna surviving and it is the permafrost is that old. the dirt that it was deposited in his presence and the horse tie. that's the only reason it survive that long. we have to know how old it is for looking up population change through time. that's a challenge. >> thank you all very much for coming. [applause].
10:40 pm
>> has 2015 wraps up, c-span presents congress, year in review, look back at the newsmaking issues, debates, and hearings that took center stage on capitol hill this year. join us thursday at 8:00 p.m. eastern as we revisit mitch mcconnell taking his position as senate majority leader. the historic address to the joint session of congress, the resignation of house speaker john boehner, and the election of paul ryan. the debate over the nuclear deal with iran, and reaction from congress on mass shootings here and abroad. good here and abroad. good control, terrorism, the rise of isis, congress year in review, on c-span, thursday 8:0. >> this new year's weekend, american history tv on c-span three has three days of feature programming. beginning friday afternoon at 310 eastern, pamela smith hill,
10:41 pm
editor pioneer girl, the annotative autobiography. discusses life of laura ingalls wilder, comparing and contrasting the tv show book series to the real-life of laura ingalls wilder. >> all wanted to write about people, places, memories that were not only to her personally but would resonate with adult readers. so, as a reviewer they pointed out pioneer girl indeed contains dark scenes of domestic abuse, love triangles gone awry and a man who lit himself on fire while drunk on whiskey. then saturday evening james swanson compares the assassinations of president abraham lincoln and john f. kennedy. highlighting similarities and differences between both tragedies. at ten p.m., the the 1965 nbc's meet the press interview with daniel moynahan, who assistant
10:42 pm
labor secretary authored a report on the causes of black poverty in the united states. >> i believe what president johnson said in his howard university speech, you cannot keep a man and change for three centuries, take the chains often say you're free to run life. people have to be given the opportunity to compete with effective resources. i believe we should make a special effort. sunday night at 930, a visit to the park in washington d.c. about a new national world war i memorial for its upcoming 100th anniversary. for a complete schedule go to c-span.org. >> , science and technology reporter john markoff on his book, machines of loving grace. the grace. the quest for common ground between humans and robots. he spoke at the computer history museum in mountain view,
10:43 pm
california. >> john markoff has been seen around the corners of the future is one of the nation's top technology writers since he joined the new york times in 1988. in 2013 he won the pulitzer prize for explanatory reporting, as prize for explanatory reporting, as part of a team of new york times reporters. is everyone host a moderator for a revolutionary series and we love john for all those reasons. but, we have a special affection for john for other reasons as well. he is a child of silicon valley. he grandpa went to high school here. he started covering he started covering technology in silicon valley in 1976. his book, what the dormouse said, illuminated the influence of the 60s counterculture of the valley on the personal revolution and he did so in a way nobody else had or has. once
10:44 pm
more he looks around the corner toward a new future of technology with his brilliant book, machines of machines of loving grace, the quest for common ground between humans and robots. if you're keen enough to catch the literary reference in the phrase, what the dormouse said, you may also understand the cultural reference of this book's title, it comes from a 1967 poem by richard brautigan which and in its entirety said this. i would like to think, it has to be of a cybernetic ecology where we are free of our labors and return to our mammals brothers and sisters and i watched over by machines of loving grace. that was brautigan's vision almost 50 years ago, we are here tonight to probe the provocative question loving grace, or something else? else? please chime in walking mean john markoff [applause].
10:45 pm
welcome. thank you. you are among friends, this is the home team [applause]. i don't know anything about this, you're the expert. i mention that you are a child of silicon valley and you have mentioned it a lot while you have been here. i just wanted to ask you, what does it mean to you to be a technology writer was so much of your personal dna coming from silicon valley? >> what does it mean to me? it took me a long time, as a kid i did not realize i was in a special place. i had no idea. actually played in the hewlett's house when eyes in first grade.
10:46 pm
virginia was a classmate all the way through. i delivered papers at the house of steve jobs, i would like to say there goes the neighborhood. i wrote my last book, dormouse because i left the day for almost a decade. it was an anti- biography. i came back and i discover there is an amazing new industry and i wanted to find out how it got there. it started as a serious in world history which i love doing. it's part of the air i breathe in a sense. i grew up with it. it's also so generational. i grew up with a particular generation and now things have moved on.
10:47 pm
the wonderful piece of research that was done about a year ago and what he did with cgl located the center silicon valley, once upon a time in the early 80s now it's at the foot of the trail hill. you can feel it, scott from a manufacturing center and it's in a different place. generationally i feel i can barely touch with it. i went up to the section in 2010, i look at it -- i grew up as a reporter it was like swimming in the sea. i was part of a group of people, now i'm distant from that which is still very real. >> so in that see, because you
10:48 pm
have seen other oceans around the world as you cover technology, before we get to the robots part of this, have you ever been able to discern what it is about silicon valley that sets it apart from so many other places that are trying to capture this mojo? >> there have been moments. personally the moment i got silicon valley, there is a point in time when i thought i understood it. it was probably a 1981, there is something called the big blue computer company -- big blue it was a computer hobbyist group -- big blue something. it was an ibm pc computer group. it mad at dyson and sunnyvale. it had the same flavor as the home group at the ibm pc was the new thing on the block. i went to a meeting, there crunch was walking around the
10:49 pm
back room, this 300 guys in white shirts and pocket protectors. there are guys. richards got up on states and she was a really a great reporter, she she basically interviewed this audience of people like you, won't she said, how many of you people want to start your own company? three quarters of the hands went up. and i said oh i get it. at that point is very clear that people felt deeply of if they had a good idea they could start a company and as part of the dna of silicon valley. there's other bits of history. i used to tell the story, why did silicon valley happen to the first point is that shockley came back because his mother was
10:50 pm
here. then there is this other point which i think is important, that that is the first at&t antitrust lawsuit, one of the deals that at&t made with the government in the 1950s was a mandatory free licensing of transistors. that hadn't happened, no silicon valley. then there's this thing and congress that allowed the creation of venture capital is inches the synergy of those three things -- i've always said that is what the valley is about. it's about learning something new from david brock who wrote the biography of gordon rower. this is in my book because he was in the archives looking through newspaper and he stumbled across a memo that shockley wrote probably in 1952. he made this case for bell labs
10:51 pm
to build something like he called the automatic trainable robot. rod brooks who started, we think of robotics baxter he just gave his employees the text of this document known could tell what the date was. he laid out the notion of whatever google might be doing now. so the connection is it makes important for silicon valley's that he went to beckman interest remits, he did not go there to make a transistor company, transistor company, he went to ask a robot company. there are robotics at the root of silicon valley. i didn't know that. he wanted beckman to build the robot size. in that kind of evolved into a transistor. the original vision was about
10:52 pm
robots. so here we are 50 years later. >> what a perfect said way into the whole discussion of robots. this book seems to have captured the national imagination. your your book tour has been extensive. you have been at a lot of places. it's doing well. what is it about our current fascination with robot has met your expertise as a writer. why do you think that is right now? >> my last book to her about driving to san jose. much smaller. so, i have a theory and i can't do it but i'll throw it out there. you know how we make fun of the japanese for being robot crazy, i actually think americans are as obsessed with robots as japanese we just do not acknowledge it. the differences
10:53 pm
we we have a love-hate relationship with robots. you see it everywhere. you you can turn around without seeing some sort of robot obsessed component, science fiction, movies, the whole thing. i think it's episodic. this has happened periodically since the invention -- initial computers, the book cyber -- there is this alarm about the arrival of automation. he had some very clear views about that. than a decade later, there is another sense of alarm in the united states and people wrote them manifesto, there's a full on government investigation on automation. the vietnam war happen and it went away because we got distracted. then of the last three or four years it has gone up again
10:54 pm
because this new way of technology that is starting to work. ai, as a field is overpromised and under delivered so many times over its history. so it's created anxiety. >> i want to get to the rise and fall of ai. right up front you say something sobering. you say how we design and overact with our intelligent machine will determine the nature of our society and our economy. it will increasingly determine every aspect of our modern world whether we live in a more or less stratified society to what it means to be human. she john, to betty then pick a book with some higher stakes. you deliver again and again in the book with that observation. >> the other day someone asked
10:55 pm
me what it is to be human? i i kind of have an answer to that question. humanity is rooted in the interaction between individuals, what is human is this thing of culture. now we are getting these machines and increasingly intelligent. as a species we talk to everything, we talked our cars, we talked our pets. as machines we are building talk back to us you're going to have to have a relationship. it is very clear that people will treat these things as autonomous beings, whether they are not. it's already happening. i wrote a story in the book about this microsoft experiment that is going on right now in
10:56 pm
china called gel ice. at the chat box. it's a chat bot. you type, and it response to you. they have gradually gotten better over time, now we are applying the internet, deep learning, big data, there is sort of an uptick. i've been dealing with since 1991, none are very believable. both siri and quintana are productivity tools. the designers of siri was to get something done. you ask a question i do get something done. gel gel ice is designed as a companion. 20,000,000 registered users, users, 10 million use it intensively, multiple conversations per day. 25% of gel ice users have set i
10:57 pm
love you too gel ice, 5050% have said thank you. it was even creeping up the microsoft designers. then i have this interesting conversation with a woman who is the former ibm researcher, she is chinese, she said you know when we come to your country it feels quiet. in china, interactions are so densely, people in contacts all the time. her her view is gel ice is private space. they call it toilet time. they go to the bathroom and have long conversations with gel ice. this is real. this is the world of her, were stepping into her. >> it sounds like her, does. as we get into this, there's another another important definition that you make early on in the book about artificial intelligence.
10:58 pm
you talk about those who believe a. i. will replace humans replace humans and those who believe a. i. will augment human capability. it's at the very root of the case. explain why it's important to understand that. >> it's a dichotomy and i'll talk about that. it's a dichotomy and paradox. there is no easy way out of it. i notice when i was writing dormouse that in the early 60s there were two labs on the side of the campus. john mccarthy created in 1962, at that time he believed building a working ai to take a decade. on the other side of campus, doug engelbert set out to build technology to augment human
10:59 pm
beings. their philosophical glia post. i struck by that because those two labs really touched on two separate communities in the computing world. the ai community and the human augmenting. i realized they to use ai technologies that are designed to augment humans as opposed to displacement. look for examples throughout the book of people who have crossed over. >> ..
11:00 pm
>> >> have we been expecting that we are human and would replace us
11:01 pm
11:02 pm
>> >> there was almost no autonomy there l all. but it's with a small amount of autonomy that you would seek to draw a circle on the screen then the machine could grab the door knob. but it that had some behavior that was autonomous. and there are technologies that are out there but is statistical with the perception is in technology but i don't think it will come as quickly as they think in they are getting
11:03 pm
increasingly familiar with that human of augmentation and questioned and to say awad autonomy was like to say but what will happen relatively soon with that social benefit. is that right? >> i continue to evolves but those are real killers that keep us from complete the selfie drive the cars for a long time. that is somebody to do -- ruth teetoo dinner i would pay for dinner. google went down that path it is almost 1 million miles
11:04 pm
and to their branch point to say there are problems that we cannot solve. to get in to a different regulatory regime and what did they get enough attention when they hit the bicyclist they may not kill them which i hear that nothing is perfect. then they ran to the handoff. but with those professional drivers and then they began experimenting at the end of
11:05 pm
a long day that was distracted behavior. that is a handoff problem you will not solve. i am sorry. there is no way anyone can solve that problem for a long time and until we are entirely at of the loop and those that did not crash. en next year we will have super cruz and largely they had this on the market already. but it's the slow speed
11:06 pm
driving system faugh where the car drives by itself up to and including comedy miles an hour then though lawyers got involved so of those technologies that will make driving safer. >> yew hedges given some statistics that 34,000 people died in car accidents and those with commercial drivers in there is that much displacement would is up with that? >> of course, all the
11:07 pm
drivers say they're still doing research for the workers but it is called the earth is a to there is a study at columbia, blood dash comparing the economics of robot)=>
11:08 pm
11:09 pm
11:10 pm
the. >> about two and half million jobs. so in jerry jobs or technical jobs it is a much more complicated subjects. if it was my reporting in
11:11 pm
2010 reporting on the fact that $35 an hour paralegal to be displaced that kicked off the frenzy. and at one point in my hair was on fire of the impact of the robotics in china. what happens when the robots come to china? the economist says he don't get it they are coming justin time. that china is a rapidly aging economy believe robots in their work force than if you're looking at china or japan in japanese society is
11:12 pm
imploding on a self. book at europe which is also aging very rapidly. >> spending $1 billion and if you look at those outtakes you don't want those robots anywhere near grandma. [laughter] we have a lot of work to do. so the people who are worried so how do explain right now in america 140 million people the labor force participation is declining.
11:13 pm
with the main street because of this technology is a small part of it. >> let me remind everybody in the audience if you want to be involved in the discussion. there is a great chapter in our book for the human race that could be a bad year what is supposed to happen? >> his view of the acceleration is that we are absolutely. i grew up in silicon valley
11:14 pm
in the they get faster faster and cheaper faster. and at this point to to be accelerating as fast as i can. end one generation where is all of that curve. in the wars lot is doubling at two years in a row so for a long time is so they went to parallelism.
11:15 pm
most recently we have slipped there is dark silicon the you cannot turn off all the transistors at the same time or it will melt third. >> furthermore that the cost of transistors that if that is true you would get to the design and with free performance and lower cost everybody but until says that is broken or has stopped. so those exponential is around the corner i am super skeptical. >> that is the year's 2045?
11:16 pm
>> data now where they are now but it was 2027. >> so i have this wonderful moment at the stanford resilience this summer in the former pentium designer was talking about these problems and at the end of of reading every added to a computer architect and was absolutely giddy to slow down to rely on the creativity is the architects turn. cool things may happen but not the way until has done 25 years. it just may become more episodic.
11:17 pm
>> that you tell about robotics where is the most wars important work being done? to spend a lot of mobile applications software was happening again d iphone happen to and i still expect at some point the will scale to though whole world. >> what is driving us?
11:18 pm
>> the dean of the information in school i cannot remember her name right now but with us in networking fax it a culture you cannot duplicate anywhere else. it is still very much alive. in red we talk about the search for a truly personal assistant to draw a distinction and then to talk about that here but the other is from microsoft. we all remember the of my -- the paper clip.
11:19 pm
end of every are and then coming from some work from stanford to believe in this concept isn't as much as people edsel levy filed the paper clip there are some things that microsoft could have done there was a cultural disconnect that undermine did it ak bader research developments hated it. they call a dead a -- clown. they hated it but it turns out the code that would make it less obnoxious there was not enough room of the distribution desk. also there was some design issues with social
11:20 pm
interaction which was fundamentally wrong. nice try but it put setback to agency a decade. [laughter] and too early that google blast from the reality. [laughter] but seriously it had a similar impact but i think to design their systems there is the tremendous amount of potential. >> apple recently said 5% speech recognition that is suppose to be but cubans do. but the system does get
11:21 pm
better over the decade as we use them. but just talky into syria is standing. i remember so with that speech recognition called the admirals' advisor. you could say left or right it may get their right to. that is in the space of 30 years to have these conversational systems that says is close to being human beings. and those of you with the give maybe better than it is. are we just being patient? >> it is coming very quickly. with those steep learning technologies that will get
11:22 pm
better quickly. >> what is holding us back? i don't think i have done enough reporting. but i can see that progress. but the point double life to make about syria and then decide that i was proud of in the book but the two architects are affected by the navigator. if you are familiar with that the shooting video that was put together that the chief visionary left the company to compete with apple so he went to alan to come up with the book and he did not think that it
11:23 pm
existed yet. in the notion which spawned it and the number but i said where is the idea? he said i was just titillate nicholas negroponte. and to i went to dick ... to save rare did you get this idea? it came from gore did. and it was this cyber-- a cyberscientists have any doubt at m.i.t. in the '80s it had a notion that human
11:24 pm
intelligence that i found a very compelling idea. that was the idea. >> the first question will be the next question i would ask that is robots as tools. you write cubans can be completely designed in that seems to be the central debate. >> i have stayed away from that and about. in the u.s. in the soviet era and then to find targets without human intervention. and that weapons system of
11:25 pm
that missile back into the arsenal. is about chided a strategic power and then to stay farther away so they designed a weapons system with that capability to fly 300 miles that is entirely out of context. the set maya autonomous weapons system. is involved in the targeting operation. to split the hairs. in that weapons systems makes that decision and intervention there are
11:26 pm
dozens of countries it is anti-radiation. if it matches the template. and i see that technology. with those devices that are a boy during overhead. but now there are too blebs hanging above washington d.c.. to do 24 hour surveillance with the cruise missile attack.
11:27 pm
but with that system in the future i don't want to do think about their. so i went to the conference and a well-known roboticist can make the case for autonomous weapons of the world's best robot designers. with that component technology they were deciding to do like the rescue profit -- robot but that is an important point. so here we designed this technology with warfare by drums.
11:28 pm
and dad is great when we have the technology. sold 80 into arms races. with small groups of dictators get the weapons. and i see those scenarios that are dark earth mbb particularly inevitable. >> now lets light said that up. sova touche driver changes in the area with the artistic and cultural forms like hollywood? >> with the two best
11:29 pm
examples of that budget to pioneer the roboticist but that may i come pretty both of them went into a i envisage the reaction that we had so as a space odyssey so with the rainbow's end about of materiality to be highly recommended then i
11:30 pm
said while. so in the books i write magically called industrial perception. and he did not go to google with the expressed intent of personal computing. if you walk down the street 50% will look down at the palm of there he and. that cannot be the end point of human evolution of. [laughter] and basically

72 Views

info Stream Only

Uploaded by TV Archive on