- I'm here today with the Science Entertainment Exchange
which is a program of the National Academy of Science.
And we're here to talk about bioethics.
What does that have to do with Comic-Con?
I have a quick story for you.
I imagine some of you have read Stephen King's The Stand.
At the beginning of that book,
there's a guy named Charles Campion
who's working as...
He's stationed as a guard at a secret government facility,
a secret military facility, in the desert in California
not far from where we are.
And it's his responsibility to make sure
that no one gets on or off the base
who's not supposed to.
And in particular, if anything goes wrong,
it's his responsibility to lock down the base
and make sure that nobody and nothing gets out.
But on the fateful day when the alarm sounds
and he realizes that there has been a containment breach
and something has gotten out,
he makes a quick decision and says,
"I can't stay here.
"I have to save my family."
And he races across the bases
and grabs his wife and child, throws them in the car,
races out of the base.
And in the process, he unleashes a supervirus on the world
that destroys 99 percent of the population.
He just made an ethical decision.
And I think we can probably all agree
that he made the wrong one
because he killed the vast majority of people
on the planet and unleashed the apocalypse.
But these sorts of decisions are the basis
for a lot of our favorite stories,
and we're gonna talk about a few of those things today.
Alongside me to discuss this are my two friends,
Mike Kalichnik.
- Kalichman.
- Kalichman, oh my God, sorry.
And Felicia Cohn.
And would you guys please tell us a little bit
about who you are and what you guys do?
- Do you wanna?
- Go ahead, Mike.
- Okay.
So I'm a scientist who got interested in ethics
some years ago, and that interest
led me to create a Center for Ethics
in Science and Technology which is directly relevant
to what we're doing here today.
My presumption is that we have some wonderful benefits
coming down the line from science and technology,
but those benefits come with risks and potential harms.
The best way to deal with that is
not just to trust a few people to address those questions
but to have everybody thinking about those issues.
And one way to do that is:
The entertainment industry provides us with
opportunities to think about those issues
before they happen and ask,
"What would I do if I was faced with
"being a security guard and a virus
"that will kill everyone and kill life as we know it
"is gonna be released?
"What should I do?"
- Right.
- I'm Felicia Cohn.
I am a bioethics director for Kaiser Permanente
in Orange County.
I am also a clinical professor at the
University of California Irvine.
My background is actually in religious ethics,
but I wanted to work in the healthcare field
so I spend most of my time in hospitals
where we deal with those questions every day.
There are a lot of things we can do.
My job is to help figure out what we should do.
- All right.
So let's dive in.
Please tweet us your questions and comments,
and we'll try to see if we can bring any of those up.
First off, let's talk about the movie Gattaca.
So Gattaca, if you guys remember,
is a movie about a near future where
genetic modification has become commonplace
to the point where the vast majority of people
on the planet now have been modified in some way
to, at the very least, filter out diseases
but also to make them stronger, more attractive,
more intelligent.
And the people who don't have those modifications
have become something of an underclass in society.
So, I guess, I'd love for you guys
to walk us through how likely is this,
how close is this, what's going on that's like this,
and what are the ethical questions
that we should be thinking about.
- I'll jump in first on the science part.
Is that we are doing genetic modification
of a variety of organisms now.
This is not just a question of a technology
that may never come to pass.
This one is there, and really the main limitations
right now are the decision not to be doing
very much of this in the human organisms.
But we know that we are genetically modifying
other species.
We've been actually genetically modifying species
for many thousands of years,
but in the laboratory now,
genetic modifications of plants and animals
are already going on.
- Well, since Mendel and his peas
who we all learned about in elementary school,
we've been making changes,
not always on purpose.
But we change our food.
We change animals to better suit our needs,
and we can do that with human beings.
And there are a lot of experiments,
a lot of research now to make changes that,
I think, we would all agree on for the better
to eliminate life-threatening diseases,
to eliminate disabilities or whatever
in our current definitions constitutes disabilities.
But just two days, the Nuffield Council of Bioethics
in the United Kingdom decided maybe it is okay
to modify human embryos in ways that
they said would benefit that person, that baby,
and we don't know what that means yet.
That might be opening a door,
might be opening a door to changing human beings
because once you change the genes,
you haven't just changed one person.
You've changed every person that that person creates.
- So can I clarify?
You said we have the ability to make these changes
about removing diseases or disabilities,
but we don't have the government authorization to do so.
Is that the...
Except now in England?
- We have some ability.
We can't do all that we would like to do.
I can't decide today that I want a perfect
blonde-haired, blue-eyed, athletic, high-IQ baby.
We don't have the ability to do that.
We do have the ability to change some things
hopefully for the better.
The problem is right now the technology
can also be harmful.
So like any other treatment,
you take too much Tylenol and it helps your head,
but then maybe it shuts down your liver
so there are side effects.
And we need to be aware of those too.
- Right.
Well, I think that there's also the questions of
what are we saying when we're talking about
eliminating disabilities?
Or for that matter, on a more basic level,
I know people already who are making choices
about what gender they want their baby to be.
What does that pose in our society,
in other cultures around the world,
the preference for one gender over another?
The challenges that that presents us.
- Well, we look at countries where
that's been a common practice,
and now they have too many men
and not enough women.
It's hard to make babies without enough women
so there are lots of implications to
every decision that we make like that.
And you know, how much farther are we gonna go?
I do clinical ethics so I deal with real people,
real questions, real requests,
and I had a couple once.
They both were deaf.
One was deaf from birth.
One was deaf by some accident that occurred later
in his life.
And they were very much part of the deaf community,
and they wanted their future children to be deaf.
And they came to our medical center asking
that their children be screened,
their embryos be screened so that they
could have a deaf baby.
So is that okay?
I think most people would say
if you wanted to screen against deafness
given our current cultural biases
that that would be acceptable.
But is it okay to do the opposite?
And what makes one wrong and one right?
- Right.
That's really interesting.
I certainly know people in the deaf community
who would say, "I am glad that I am deaf.
"My family chose against giving me a cochlear implant."
The deaf community is an active culture.
I actually had the chance recently in Tel Aviv
to participate in a activity at a museum
that gives you the experience of being deaf
for a limited period of time,
and it was really interesting to
kind of rely on your other senses
and see what a different life experience that is.
But that's fascinating to know
that it's a choice sort of after birth is one thing.
For people to choose to have a deaf child
is something I haven't heard of before.
That's really interesting.
- Well, fortunately the question at the time was easy
because we didn't have the technology
to actually make that happen
so we could say no in good conscience.
But as technology progresses, what will we say?
- Also to say to the extent that this can
be genetically determined and we can make that choice,
people might want to.
And for those listening into this podcast
who are hearing, you might find this appalling
to think that somebody would choose
for their child to be deaf.
But think of it another way.
Imagine that you could not easily communicate
with your children in the language you speak.
And the deaf community has a way of communicating
that, if you learn from birth,
you're very comfortable with it.
You have the nuance.
You communicate well.
What happens if you pull that apart?
You've got two deaf parents and a child
who's being brought up in the hearing community,
and that child is able to communicate
better in that world than in your world.
And that's something that many of us
could well understand why they would choose that.
And so the question we're going to be asking today
for all of these issues is:
What should we do?
And who gets to decide?
Is this something that each individual
gets to decide for himself or herself?
Or is this something that we want society to regulate?
And if so, what do those regulations look like?
- Right.
What other sorts of...
What are the things that we know are...
What do we assume is a given
that we want to screen against?
What are the obvious...
These are the things that we are now...
What is in practice?
- You know this probably better than I do.
- No, I don't actually know.
- Oh, okay, got it.
Well, I would assume...
We talked about the idea of certain diseases
that pretty much make it impossible
for a child to grow up.
- I think that we probably aren't in a position
to give you a list of things,
but one of the keys here is that
most characteristics of people are determined
not just by one gene
so you might have to look for several.
But let's say that two people, a couple,
decide to have a child.
They have embryos made in vitro, in the laboratory,
and then they want to choose among those embryos.
So they could test those embryos genetically,
and one example of something that they
definitely would want to remove
would be the gene that determines having
Huntington's Corea.
Specific disease.
A disease which you don't see any evidence of
until, I believe, you're in your 30's or 40's.
But at that point, it's a difficult, horrific disease
from which you will likely die early.
That would be something that you wouldn't wish
on your future child
so you choose an embryo that
doesn't have that characteristics.
Very different than saying,
"I'm going to go into this embryo
"and choose genes that are gonna make this child
"taller, or have blonde hair."
- Right.
And part of why we can't answer that question
is because, in most countries, testing
or experimenting on human embryos is not allowed.
It's specifically illegal in a number of countries,
and other countries just don't speak to it at all.
So it just hasn't happened
which is why that report out of the United Kingdom
was so stunning because it might start happening now.
And what is that gonna mean?
And we already are doing things like
what Mike was talking about.
We have couples who have a child with a serious illness
who needs maybe an organ transplant
or bone marrow transplant,
and they're having trouble finding a matching donor.
And so they will conceive another child
through in vitro fertilization using
pre-implantation genetic diagnosis
to determine what the likelihood
of each of those embryos of having that disease are
or of being a genetic match
to their already existing child.
Then they will implant that embryo
that is the best match, that is disease free,
at least that disease free,
and bear that child and then take parts of that child
whether it be stem cells from the umbilical cord
or bone marrow or, as the child gets older, a kidney
or part of a liver to implant
or to put into the other child who's been ill.
So you're giving birth to one child
to save another child.
- Which actually kind of moves us into our next topic.
So I'll reference...
We've got two movies,
actually three, off the top of my head
that touch this subject of the idea of
creating a life to sustain another life.
I think first of Never Let Me Go
and this group of people who have lived...
I remember, you're trying to figure out at first
what actually is wrong with these people,
what's special about their experience.
When you realize that basically these are clones
who've been created to provide organs to other people
and that their entire life is controlled.
And yet, they're still people with emotions
who want to be able to live full, meaningful lives.
The Island is another that did that.
And My Sister's Keeper is the less sci-fi.
And actually probably, I would imagine
there are true stories that are identical to that.
- If less twists and turns.
- Probably with a few less twists and turns.
But you never know.
So yeah, let's talk about the morality of that,
the ethics of that:
of creating a life to sustain another life.
You wanna give us a bit of a...
- Well, it's human identity.
We have children because we want a child.
And here the purpose for having a child
is to save another child
so what is that second child's identity?
And what if it fails?
If I'm born to give my bone marrow to my older sister
and she dies anyway, and now my parents are stuck with me,
and I haven't saved my sister who they already loved,
how are they gonna feel about me?
I'd like to think they still love that child.
And so far, I think, in the situations where
this has happened, it hasn't been a detriment
to that child.
But I have to wonder about growing up
knowing that you were born to save someone
and you couldn't do that.
- Is this something that's actually like
relatively commonplace?
This happen a lot?
- [Felicia] No, not a lot.
- Okay.
The thing that I thought was particularly interesting
about the example of My Sister's Keeper
is the idea of:
At a certain point, that child is gonna know
why they were created and what
the life is that they have to look forward to
at least until their sibling is healthy.
So at what...
Ultimately, when does a person get to take control
of their own health and well being
and make the decision of,
"I don't wanna have these procedures anymore.
"I love my sister, but it can't be my obligation to...?"
- Right.
"I love my sister, but I don't want surgery."
- Yeah.
- "I'm scared."
Or, "What if I need that kidney later?"
- And shouldn't, I guess, the parents
are ultimately...
The parents are the ones making these decisions
up until there is resistance to it.
I would assume you must have doctors
who are looking at for both patients separately.
- There are.
There are different doctors.
For any transplant procedure, there are
always two different teams.
So there isn't that conflict of interest.
But there's still gonna be overlap,
and as the child gets older...
It's easy to consent for an infant or a toddler,
but by the time kids are four or five...
Anyone who's a parent knows,
they have very strong wills of their own.
How do you?
Getting a child to take antibiotics
when they don't want to take the
yucky-tasting medicine is hard enough.
Getting them into an operating room
against a child's free will...
Even if they're not an adult,
they don't legally have any rights to make decisions.
But they still have their own beliefs,
feelings, desires.
To go against that and then still have to raise that child,
how do you do that?
- Now to be clear.
Also in our legal system in this country,
that child, even though they may not
be able to make a lot of legal decisions for themselves,
in almost setting would be asked to give assent.
And they could say no.
This is...
This and the example that Felicia gave of
maybe it won't work are reminders of
what we are looking at when we
look at cases like this.
We are almost never looking at certainty.
We're not almost ever able to say,
"If you do X, this is what's gonna happen."
If you would, a lot of ethical discussions
would be a lot easier.
So in a world where we have uncertainty,
what kinds of decisions do we want to make?
Who gets to make those decisions?
What value do we see of creating a human life?
What does that mean, in this case,
that we've actually chosen to have a human being
for a particular outcome that may or may not
ever occur?
And should we be allowed to do that?
You protect against that if you say
you're not allowed to do that.
- I have, sort of, a very sci-fi proposal,
but I feel like this must be discussed.
But I assume research is being done to
try to create those, create organs, create bone marrow,
create those things independent of an entire human
so that you could just create
the specific item that you need.
Or for that matter, to create a human
that doesn't have any sort of brain.
Tell me what that brings up for you.
- So it's certainly a direction we can go in.
And some pieces of that are already being worked on.
But it is extraordinarily scary
figuring out how we're going to draw those lines.
Personally, I feel pretty comfortable saying,
"If you could grow a kidney for me in vitro,
"just a kidney that is perfectly matched for me,
"and replaced my defective kidney,
"I feel that sounds fine."
If you go to that level that you just described,
maybe create an entire human but with no brain,
who's going to start drawing the distinctions
because you will need some level of brain
for basic bodily function?
How's it going to breathe?
So you say, "Okay, well, we'll have a brain stem
"and autonomic nervous system, various things."
So you can say, "We're not gonna have cortical functions,"
but there are various things you might need.
I mean, things that we do normally.
And this is why, if you have dementia,
you may die because you aren't able to do
normal functions that would help protect you
from fluid accumulation in your lungs, for example.
So how much of a brain do we give someone?
And so you could start saying,
"Well, we could have almost an entire brain,
"but we just drug them so that they aren't aware
"of what's going on."
And as you move that line further and further back,
suddenly you're creating human beings,
real human beings for...
- [Gillian] What are the restrictions on doing that today?
- Well, it's just not done.
And we're still arguing over what constitutes
a human being and the death of a human being,
more specifically.
So if you're creating a human being who has no brain
or less of a brain, are they still a human being?
Or are they dead?
And then we start getting into some of these other
apocalyptic, Walking Dead kind of scenarios:
People who aren't really alive
but also aren't really dead.
The definition of brain death is something
we thought was a sort of settled ethical and medical issue,
but it's not.
We had the case, in the last few years,
of Jahi McMath who was in the news constantly
declared dead by neurologic criteria, brain dead,
but her parents argued she was still alive.
And in some states, New York, New Jersey,
there are exemptions in the law
where if you don't accept brain death
than you're not dead.
So there's a death certificate in California,
but she was still alive in New Jersey.
And could you take her organs?
If she's dead and still on machines,
could we have harvested her organs
if her parents didn't believe she was dead?
What would that have meant?
And could we bury her if she's still breathing?
- Well, hopefully, she wouldn't still be breathing
when she's buried.
That's a...
- And breathing may not be your only criterion
of what it means to be alive.
What is one of the wonderful things about science
is it's giving us a window now
into things we couldn't have anticipated before.
It's giving us opportunities we couldn't have
anticipated before.
One of them is to even harvest organs in the first place.
But now the question of when you're going to
use someone's organs is extraordinarily difficult.
There is a group in Canada,
and there may be others as well,
that has been using a technique called
functional magnetic resonance imagine
to look at the brains for people
who we believe are in effectively a vegetative state.
They will never recover function.
And using that technique, asking them questions
and setting it up so they believe
they can get...
That that person, if they have any control of their brain,
can actually answer those questions
and discovering that some of these people
who we thought are in a vegetative state
are actually able to consciously answer
things they have heard
and come up with a response to that.
I'm not guaranteeing that those people
really have function in a way we think of function,
but it again is opening up a window into something
that they wouldn't have even considered
just 20 years ago.
- Okay.
I feel like we should move onto a slightly more...
This is a fun topic for me,
but we talked about Minority Report
and the idea of being able to predict
future bad behavior.
And as science develops, we're learning how to
look for markers in people's genetics
and also in other aspects of their anatomy
that can indicate that they might have
a future inclination towards criminal activity.
So can you give us...
Can you tell us a little bit more about this?
- Well, first, I have to remind people
in Minority Report, I believe, the terminology was precogs.
- Yeah, this is a little different than an actual precog.
- So we're not talking about this in the real world.
But just the idea was intriguing
that with these precogs, these individuals
that are basically nurtured in order to
predict the future, what's gonna happen with an individual,
we have no mechanism in science that I know of
that's actually gonna work on that.
But the idea is intriguing
that you could somehow predict whether
an individual is going to, for example, commit a murder.
And so the presumption in this movie,
based on a novel from Phillip K. Dick,
or a story from Phillip K. Dick,
is that you could stop somebody before
they committed the crime because
you know the crime they're gonna commit.
First thing to point out is the same thing
I mentioned earlier: most things we deal with
our matters of probability, not certainty.
- And that's ultimately sort of much of the plot
of that movie has to do with flaws in that system.
- Exactly.
So now let's translate that into something we can do,
and one of the things we can do is
we can measure neurotransmitters in people's brains.
We can look at brain anatomy.
We can measure brain function under certain circumstances.
In every one of those cases, we might ask questions
about what are the characteristics of somebody
who would do X?
What are the characteristics of somebody
who would commit a crime?
So what if we found that 95 percent of people
who have this feature are likely to go on
to a life of crime?
What should we do with that information?
Should we sequester them, quarantine them?
- But what about the five percent of those people
who don't?
I think that the...
We talked a little bit about the James Fallon Ted Talk
which I watched honestly more than once
because I have a certain interest in sociopaths.
They make great subjects for movies.
So the...
As I recall that, and I think you guys might remember
other elements of this talk,
but he had done a series of PET scans on brains
of sociopaths and had found certain patterns
or elements that were consistent among sociopaths.
But I also remember that reading somewhere,
or I don't remember if this was part of his talk,
but that not all people who have that brain
are going to become violent serial killers.
All people who have that brain aren't gonna
become criminals.
They might be perfectly functional members of society,
and we would never know that they have this problem.
And I remember there was something about
a trauma in childhood is what kind of
flips the switch for many of these people,
that turns them from relatively ordinary
members of society into mass murderers.
But that's probably a small percentage of people
who have that kind of brain.
- Well, so the first question is:
How many have that pattern?
And I remember the work, but I don't
remember the percentage.
I think it is a small percentage.
I think you're right.
The idea that certain triggers
are necessary along with this makes some sense,
but again, since you don't know whether
somebody's going to get that trigger.
And a really good question is:
What is a trigger for somebody?
For somebody who has lived a life of ease
and has everything brought to them anytime,
a trigger might be that they can't
see the movie that they want to see today.
And for somebody else, the trigger might be
that they have no food and they are
at risk of being killed in order to try and survive.
So that's not something that we can predict.
It's not something we can say with certainty,
"This is a factor that's gonna lead to this
"possible outcome."
So what, as a society, do we wanna do with that?
It's kind of knowledge that
you might not be able to act on
or your might not want to act on.
- Well, I think it's similar to medicine.
We all have, within us, certain genes
that if triggered could result in disease.
So do I go and presumptively have
certain of my body parts removed
as you do if you have the breast cancer gene.
You don't have breast cancer yet,
but do you treat it as if you're going to have it?
Do you make that assumption when we really don't know?
And lots of diseases are like that.
- Yeah, although I think what's interesting,
and this goes back to the question of
who makes these decisions, but when you have
that situation, it's my choice
what I decide to do with that information.
But if the government has the ability.
I'm saying the government with like a large G.
Has the ability to measure this in people,
does this become sort of standard testing
at some point in time?
When you're born or before you're born,
are there tests that can be done
to show whether or not you have the likelihood
of becoming a criminal?
And if you have that likelihood,
what kind of restriction?
I mean, that's the...
What kind of restriction do you put on that life?
Can you ever?
Do you just have people monitoring them?
Should we be implementing any sort of
system to test our children for these things
as they're going through school?
- Well, we do in other ways.
We do drug testing.
As a requirement for my current job,
I had to undergo drug testing.
So what would it have meant
if I had a positive result and claimed,
"But I'm not a drug user."
What does that mean?
And what restrictions would that mean for me?
And we do that with other illnesses too.
Tuberculosis: You test positive for tuberculosis,
they're gonna start you on med...
The government is gonna require you to be quarantined,
start you on medication so it doesn't result
in a public health threat, an outbreak.
But what if you're not symptomatic?
What if you're fine and now they've told you
for six months you can't leave your house
and you have to take this medication?
- Right.
And I feel to some degree that the
answer to these questions probably changes over time,
and it depends on the culture we're living in
at the moment.
The question of what sort of tolerance do we have
for locking up innocent people,
people who would never do something like this?
And what tolerance do we have, for that matter,
in locking up people who have never done anything
but are highly likely to?
- Well, in all those cases, again,
this probability issue.
We already make these kinds of decisions in society.
We make choices about what is the risk
that if I lock this person up that I am
wrong in doing so?
And you could say, "Well, we should never lock up
"anybody who isn't guilty,
"and we should only lock up people who are guilty."
But just look at the way...
- But by whose definition?
- But just look at the way courts define guilt.
You hear terms like beyond a reasonable doubt,
preponderance of the evidence.
You don't hear absolute certainty
because you almost never have absolute certainly.
And I think if you would go into society,
ask the people who are listening into this discussion,
there are a lot who will say,
"I would rather err on the side of making a mistake
"and not imprisoning somebody
"because I don't want innocent people locked up."
- Right.
That's certainly my leaning.
- Yeah, 'cause I don't wanna be locked up personally.
But there are others who would say,
"I am so fearful of what might happen
"because of some criminal out there
"that I would rather lock up anybody who
"might be a criminal."
This is a question of society.
You mentioned different societies might have
different questions.
A society makes its decisions.
We make those decisions by our votes,
by our input for how we think we should operate.
- And the same society may change over time.
We've locked up Japanese people during the war
because they were Japanese people.
And now we think that's wrong.
A lot of people probably thought it was wrong then as well
yet it happened.
- Yeah.
Well, we're moving through these kind of quickly.
- Okay, wait, we've got plenty...
So maybe sort of as a aside to this topic
is if you're talking about predicting behavior,
the question is: What can we see in a human
that tells us what their inclination is,
what they might do?
And one side topic that we were considering talking about
might be worth considering is:
What about lie detection?
So what if you could look into somebody's brain
and tell if they were lying or not?
And at the risk of doing some advertising,
but actually I don't know if it's still
the name of the company,
there is a company in San Diego that...
I won't give their name because
I don't remember the name for sure.
But the company that uses MRI
for detecting whether someone is lying or not.
There's been a lot of work done on this.
You put somebody in this magnetic resonance image machine,
usually most of us think of it as
looking for brain abnormalities or something,
for example, if you're looking at the brain.
In this case, using it to look at function
in different areas of the brain
and seeing which areas of the brain
are changed when somebody's telling the truth
versus when they are being deceptive.
And on the surface, just as I asked,
it sounds like, "well, that'd be easy to do.
"Just put people in when they're lying
"and people in when they're telling the truth."
But I need to point out that that's not easy.
We can come back to why it's not easy,
but let's assume you thought you had done that
with a reasonable level of confidence.
Now you stick somebody in the machine
and by having them in this machine,
you can ask them a question
and find out whether they're lying.
So what if we could have that machine at the airport?
Everybody going through security...
- Well, this the question I had:
How much of this was practical to implement
without putting somebody in a machine?
- I will stipulate right now that
if you think lines are long at security
at the airport now, imagine trying to
put people into a machine where you have to
repeatedly ask them the questions.
And basically nobody would ever fly.
But that's the kind of...
Every technology starts out very expensive,
very slow, very cumbersome.
I mean, look at the cellphones we carry around now
and compare that to...
- To the mainframes that took up the size of this room.
And that's just in our lifetimes.
- Right.
So possibly in the not-distant future,
we'll have...
- Fingerprint scanners for lying.
- In terms of the distance in the future,
there was an interesting story I saw
a few years ago that said
that whenever somebody's in a panel like this one
and somebody asks them, "Well, when will this be?"
They always say, "This is five years away."
Five years is sort of the perfect point.
Far enough to say, "Okay, well I know there's
"still work to be done."
But it's soon enough and sounds,
"Oh, this is imminent."
The real answer is you don't know
because people are working on these questions
from all different directions,
and somebody might have an insight, a lucky breakthrough,
and suddenly the impossible becomes commonplace.
- And affordably.
- And affordably.
Well, actually this is a really good lead in
to our next topic which
I think the best movie example we had of this was I, Robot.
In I, Robot, we have robots that are now
a part of our every day lives,
and they have been coded with a series of principles,
the primary of which is you can not hurt any human.
And we see today that we have AI
around us in a lot of different ways,
becoming increasingly common,
and we are entrusting these systems
to make decisions about our wellbeing.
So driverless cars being the topic
that we started discussing.
What do we do when there is a situation
where there's a person outside the car
who is at risk and there's a person
inside the car who's at risk?
And we are leaving it in the hands of AI
to make the decision of which life is more important.
So please tell us more about this.
- I think the question there is:
What algorithm would you program into the car
to make that decision
which is still gonna be human-based?
So it's still gonna be human biases guiding a machine
at least until AI truly becomes I.
Human machines truly become intelligent.
But to do that the machine would have to
experience emotions, and that's something
that I don't think anyone has figured out.
I mean, there are humans who don't experience
enough emotion, much less machines.
So how would you calculate that?
And then how would the machine have any way
of knowing who that person behind their wheel is
which would be easier for them to know
than who that person outside of the car is
to make that value judgment.
And then what is it based on?
And that reminds of when we first started
doing kidney dialysis.
The machines were enormous and expensive,
and a hospital may have just one or two.
And so much need.
So many people in kidney failure who would benefit,
whose lives would be extended by access to this machine.
So they had to make those decisions.
And how do you do that?
- Right.
And to some degree, that exists now
in terms of organ donation and
who is the recipient of those organs.
"Here are the criteria that we use to decide
"who's at the top of that list."
- Yes.
But at the time, Congress couldn't decide.
Couldn't decide and they were the deciders
because they were funding it.
So they made it available to everyone.
So it is probably the only service
that is 100 percent funded by the government.
Your kidneys fail, you will get dialysis
and it will be paid for.
- Oh, that's interesting.
- Because we couldn't decide.
But at the time, they had groups of people
who needed this.
So maybe it's easy to say, "Well, we'll do the children.
"We'll make sure the children get this first
"because they're young and innocent
"and deserve it, deserve a chance to grow up."
But then what if there's a physician among that group
who could help save some of those children,
who's doing all his research on kidney transplantation.
He could save these children
so maybe you save him.
And them maybe there's the mother who has five children,
and maybe you need to save her
so that those children don't become orphans.
But then we get into the harder questions
where what if the person who needs it is a prisoner,
a current inmate?
They're still a human being.
Are they worthy?
And so there were committees set up to do this
that made social judgments.
And that, as you can imagine, became problematic
so it became medical criteria.
But even in the medical criteria,
bias sneaks in because you have to have...
For a transplant, for example,
you have to have not been drinking
if you want a liver transplant.
You have to have demonstrated that you can abstain
for six months or more.
So is that a social judgment or a medical judgment?
Or both?
And you have to have a support system
so if you're homeless and don't have family
to help you make sure you take your meds every day
which seems like a reasonable criteria.
You have to be compliant so we don't waste the organ,
but is that really a social judgment
because you're homeless and don't have loved ones
who can help you?
- So the kidney example is a really good example
of what we would do with perfect information
and even with, what I'm gonna call quote unquote
perfect information, is still tremendously complex
and value-based and judgements may vary.
But when you or I drive a car,
much less when a car that is autonomous
and making decisions on its own,
in all of those cases, we don't have the nuance
of whether that person standing on the sidewalk
that we would have to hit in order to avoid
hitting that other car,
we don't know whether they're a PhD scientist,
whether they are a drug addict,
whether they are the president of the Untied States,
we don't know who they are,
and so the car is now having to make decisions
in other ways.
- What I'm starting to picture though is
as facial recognition technology develops
and we're programing AI to make these sorts of decisions,
we're gonna get to a point where
the car is going to be able to know,
"Ah, that person out there who is a doctor
"who's doing important research,
"and this person in this car who has
"the brain pattern of a sociopath,
"I'm gonna choose to save the guy out there."
I mean, that feels like that's actually
not that far in the future.
- It's a good example,
but what I think is more likely
is not that that will rely on face recognition,
but I can easily envision a future
in which people allow themselves, if not are forced,
to be identified with some sort of tag
so that that tag would be providing information
to let the car know.
Again, for those listening, they might ask themselves
at this point, "So if I'm in my autonomous car,
"do I want it to protect me no matter what?"
That's the first choice.
And I think most of us start with that assumption.
I'm the driver.
Of course I would want that.
But then if you step back and say,
"What do I want the society to do?"
You'd say, "Well, if the car could choose
"saving one person or saving three people,
"the car might hit three people otherwise."
You'd say, "Well, save the three people,
"and that one person should die."
Then you start getting into the nuance of
if you could know who they were.
That's probably the question I think furthest out,
and the question for all of these
that we should start with every one of these questions
is the science.
What can we actually do?
How good is the recognition,
the visual recognition a car can do of its environment?
And that's what we're still working on.
- But if you start with that assumption,
like in I, Robot, where the artificial intelligence
can not harm a human being
then could the machine even work?
If it's put into a situation where someone
is gonna be harmed.
- [Gillian] You would sort of have a shut down effect
if there is no option...
- That wouldn't harm a human being.
- I forgot the exact wording,
but I think the wording was such that
if you had to choose between two human beings and one,
you would choose losing one life instead of two,
things of that sort.
- But that still is a human bias
that we are presupposing one really philosophical theory
is better than another,
that life doesn't have absolute value,
that it's all relative.
- Yeah, so one of the things that's really intriguing
is to try and dig deep and think,
"What kind of level of intelligence are we looking for?"
We think of ourselves as a gold standard,
and we're pretty bad.
So now imagine a robot, a device,
an artificially intelligent device we've created,
and can it get the nuance?
So we have Siri now and other devices
where we can talk to them and ask them for something.
What if they give us information
that would cause us to lose our life?
But they have no context to know that
because they told you, "Turn right here,"
that a meteor's gonna hit there.
So they can't always make the decision
that would protect life.
And we should not make the assumption
that our technology will do any better
at this than we can.
And we don't do a very good job of it.
- I'm noticing in the chat right now
that there's a discussion that's happening
about what we were talking about earlier
with regard to harvesting organs to use on another child.
I think we should clear up some confusion here
because they said, "Wait, they kill the other kid?"
"If they need to, yes."
Just to be clear: We're not actually talking about that.
- Let's have a vote on that.
I think all three of us are in favor of not killing.
- All three of us are in favor of not killing anybody.
But also, we are not having one child
specifically to give their organs to another.
- So that might be the premise of that movie,
Never Let Me Go, but it was not the real life.
- And actually those aren't children.
Those are clones.
Those are rich people paying for duplicates of themselves
as backups.
That's also fiction.
So just to be clear.
- Which is not yet possible for humans.
Cloning, not yet possible for humans.
- For a kidney, which you can lose one kidney
and still be absolutely fine in most cases,
there have been discussions.
I don't know whether...
And Felicia may know how often that's
actually come to fruition, if ever
where families would say,
"There is a known risk of somebody needing a kidney
"and to have a good match.
"If we had another child and that child would
"be able to donate a kidney."
It would not be to kill that child to take that kidney.
But the presumption would be that child
would be able to give up their kidney
to someone else.
- And we do that now.
We don't necessarily procreate to create a donor,
but there are living donors who opt
to give up a kidney or part of a liver
to save usually a family member
but sometimes a stranger.
- You know it's interesting.
We have five minutes left,
and I know there's a subject I wanted to come back to
for our wrap up.
But the one thing I was gonna say,
specifically about kidneys, because it occurred to me
after we stopped talking about that.
I read about a case years ago about, in fact,
a serial killer, a very serious serial killer,
who found out that he was a match
for some, I think it was the brother
of an ex-girlfriend or something like that
who needed a kidney donated,
and this inmate wanted to be able to donate a kidney.
And it became a big court case
of whether or not the court would allow him
to donate that kidney
or whether he's being coerced into doing that.
- Prisoners are actually in a special category
in our population.
We think of them as just being restricted,
but they are actually protected in specific ways
because of their lack of freedom.
So there is that concern that they're being coerced,
that maybe they'll get a shorter sentence,
maybe there's some benefit to that if they will donate.
And then I was just reminded of a movie,
and I can't recall what it's called,
that Will Smith starred in
that he wanted to become an organ donor.
And he set about finding all of his recipients.
- Yeah, I think it was Seven Pounds.
Is that what it was called?
- I think so.
And then he went about killing himself
in a particular way so that his organs
would still be viable,
and he had figured out who he wanted
his liver, his kidney, his heart, his lungs to go to.
You have to question:
What's the motivation behind that?
If you're gonna be a living, unrelated donor,
you're not doing this to save your brother
or your mother, your child,
why are you donating?
And I see somebody saying, "Kidneys for sale."
It's still illegal.
- Still illegal to sell a kidney.
Still illegal to buy a kidney.
- So all of this, it really comes back to
that question again of entertainment we can get from it
'cause that's why we're here.
You go to a science fiction movie
and maybe see things blown up
and people running in chases, whatever,
but those allow us to think ahead of time
about challenges that we may actually
be facing already or which we may face in the future.
And being able to do that is a valuable thing
so we can pat ourselves on the back and say,
"There's good reason to go and watch that
"next blockbuster."
- Yeah, TV shows, movies, comic books,
they allow us to play these scenarios out
so we can see what is the end.
What are the possibilities?
And right now we have the latitude in the media
to play it out in ways that really aren't possible,
but you never know where the mind will go.
- Right.
Well, and it's exciting to see
how we feel about subjects before they're put
into action in the world
and what would we do if we were in that situation.
So, yeah, I was asked earlier today
why it is that as a film person
I am participating in this science organization,
and it's because this subject matter to me
is so fascinating becomes the basis for
so many of my favorite stories.
So it's really worth while to kind of expose ourselves
to these different ideas.
- Yeah, exchange is so helpful too
because it gives us the chance to
talk to scientists, to talk to entertainment people,
have them connect so that there is some reality
in the science fiction we get to see and think about.
- Right.
I certainly like that.
I think that's about all we have time for today.
So I have no idea who your next panel
up here is today, but you should stay tuned
and enjoy what they've got for you.
- So thanks.
- Thank you.
For more infomation >> আনুশকা শর্মা সম্পর্কে এই তথ্যগুলো জানেন কি? (Did you know these info about Anushka Sharma?) - Duration: 4:34. 
For more infomation >> Barry White - Just The Way You Are (Music Vídeo BR) - Duration: 6:59. 
For more infomation >> Funny Farm | Just Joking—LOL - Duration: 1:25. 
Không có nhận xét nào:
Đăng nhận xét