Gossett’s used to be humid when you ask
what’s universal to humankind people
have claimed all sorts of lofty things
music language reason mastery of fire
and cooking in cuisine or even opposable
thumbs
but sadly enough what’s most universal
of all might be gossips every single
culture in the history of humankind has
been plagued by gossips humans religions
through the ages have proclaimed
stricture after stricture against gossip
Christianity in the New Testament Islam
in the Quran Buddhism in the Pali Canon
it goes on and on which tells you how
bad the problem really is but at least
gossips used to be human I’ve been
watching our societies become
increasingly populated by nonhumans over
the past generation populated by
artificial intelligences from a trickle
of AIS in the pioneering days to mass
immigration into our societies today
it’s a tectonic shift that our
imaginations still struggle to come to
grips with and I’m guilty of having been
a part of that tectonic shift because
back when I was a young PhD student AI
was dominated by rule-based systems
systems based on mathematical logic and
I fought with my advisor for six years
he gave me a tiny little problem he said
look
we’ve built one of the first dialog
systems you know the sort of ancestors
to Siri and so on and we just have this
little problem the user comes up and
says how do I cut the disk space and we
can tell from the syntactic parser that
the verb is cut and the direct object is
disk space but the problem is when we
try to plug it into the next step the
semantic interpreter it’s got to tell
you what is that next step is it cut as
in cut bread so meaning slice or cut the
grass meaning shorten or cut class
meaning skip or cut the cheese there are
so many senses you just have to go and
write a little module that will figure
out which sense of cut that is do that
over the summer so I went away less than
one week I came back to him and I said
there is no way on earth you can solve
this with logic because this is all
shades of gray this is all ambiguity
this depends on the context this is the
sort of thing that we do so well which
is not based on logic it is not based on
reason it is based on our intuitive
abilities to integrate shades of grey
from many sources and to make that our
own subjective decision and that is the
root of human intelligence and it took
six years of screaming fights not me
screaming my advisor screaming and
finally six months before I graduated he
said to me one day after I’d given him
yet another draft of a chapter he said
wow guy this is the most amazing thing
ever I finally understand whether you’ve
been saying all these years and that
made me the happiest young student ever
and by that time I had been recruited to
go with 400 other faculty who came from
UC Berkeley where I was and Stanford and
MIT and Toronto and Oxford and Cambridge
to go found a super ambitious new
research university in the American
style
Hong Kong which is a remarkable story
that I’d love to share with you over a
beer Sunday but when I arrived I found
myself thinking where should I apply
this new paradigm that I’d been fighting
for to make machines learn by themselves
from big data and I decided to make them
learn how to translate between languages
because Hong Kong like many other places
is struggling with a multilingual
multicultural problem and so
specifically I made machines learn how
to translate between English and Chinese
for the first time and this this is how
I met one of the creators of artificial
gossips that has shaken the news sphere
many of you have probably heard about a
company called Cambridge analytical this
company worked to use AI to gather and
spread information on a massive scale so
as to influence the outcome of voting in
numerous countries including the US and
the UK and what eventually emerged is
the news that Cambridge Anika was backed
by a reclusive brilliant hedge fund
billionaire named Robert Mercer but
before he took those Wall Street offers
20-some years earlier that made him a
billionaire Mercer was a computer
scientist at IBM Watson working on human
language technology or natural language
processing and specifically he was
working on getting machines to learn to
translate between English and French the
Association for computational
linguistics is the main International
Organization for natural language
processing and about six years ago they
decided to establish an honour called
the ACL fellows in its history of over
half a century the field had never had
anything like nobel or fields or fellow
honours so to catch up with that they
decided that on this one
historic occasion they would name 17
founding fellows Danford breakthroughs
that led to a eyes like google
translator microsoft translator or yahoo
translate the two founding fellows they
named where mercer and me which is why
I’ve been struggling and thinking a lot
about the consequences of the work that
we do developing artificial intelligence
and machine learning those consequences
are enormous our societies today already
are no longer exclusively human in a few
short years a eyes have become our
secretaries our navigators our dictation
assistants our translators our shoppers
our matchmakers our curators our
journalists and our news Clippers our
house cleaners our chauffeurs our hedge
fund traders our deejays our customer
service representatives and our gossips
all our cultures and religions have
warned through the ages against idle
chatter quarrels conflicts gossip
slander condemnation calumny all idle
chatter and now we’ve gone and created
chatter BOTS artificial gossips no
gossips do two main things they hear and
they speak maybe you know this word
quidnunc this means a person who seeks
to know all the latest news or gossip in
other words a busybody this is the type
of gossip who likes to hear and on the
other hand you have the gossip monger
this is a person who habitually passes
on confidential information or spreads
rumors this is the gossip who likes to
well artificial gossips are
exponentially dangerous regardless of
whether the gossip is true or false when
gossip is false or private artificial
gossip mongers spread fake news and
confidential information like human
gossip mongers – but exponentially more
dangerously Douglas Adams in the
Hitchhiker’s Guide to the galaxy
observed that nothing travels faster
than the speed of light with the
possible exception of bad news which
obeys its own special laws well now we
have artificial gossip mongers to give
gossip even more exponential reach and
speed gossiping fake news and
confidential information are powerful
ways to change society’s views even in
the absence of evidence and it doesn’t
take that much there’s a recent study
from UPenn and the City University of
London and suggest that even if only
roughly a 25% minority of a group starts
to spread an opinion that is enough to
tip the majority to switch opinions and
in this age of botnets artificial
gossips easily account for more than 25%
of our gossips artificial gossips push
us past the tipping point for fake news
to change society’s views and just like
humans artificial gossip mongers
weaponize the power of suggestion just
like human gossips they spread
unevaluated claims that are based on
superficial appearance they promote
hearsay rather than evidence they
promote stereotyping as an old Korean
proverb says words have no wings but
they can fly a thousand miles and even
when gossip is true artificial quidnunc
gain social power just like human quit
knocks – but exponentially more
dangerously
in the psychology of rumors which is a
social psychology classic from 1947
gordon allport and leo postman point out
how gossip serves to create a sense of
cohesion among various groups of people
but to take a divisive stance against
someone else gossip actually becomes a
social control mechanism that gives
power to whoever does it
gossiping releases endorphins and gives
pleasure propagating gossips helps them
escape from an unpleasant routine escape
from negative feelings or stress gossips
become the center of attention for the
folks who are receptive to their rumors
if you think Game of Thrones Varys is
scary our artificial critics are
potentially far more terrifying masters
of whispers Cambridge analytic afore
example was an artificial quit no
Facebook is an artificial quit them from
which gossip can unintentionally leak
just as happens with most human
equipments it’s hard to keep a secret
and in this case leaking from one
artificial quidnunc to others including
Cambridge analytic oh but Siri Alexa
Cortana Google assistant they’re all
artificial quit notes and again it can
be hard to avoid leaking what you
overhear even Alexa recently
accidentally spread a private
conversation that it overheard in a home
whether our AIS are hearing gossip or
speaking gossip whether their artificial
Quinn monks or artificial gossip mongers
we exist already today in a tangled
Society of human and artificial gossips
so here’s the question I want to ask
each of you have you joined the
artificial gossips are you an unwitting
part
of the network of the little birds have
you joined the artificial gossips in
creating divisions rather than bridging
them gossip ostracizes persons or groups
exponential artificial gossip
disruptively ostracizes persons or
groups exponential divisive ‘no smiie
–nz that we face the threat the
existential threat of an extinction
event what are each of us doing about
our own habits to prevent the
elimination of our civilization you may
have heard the ancient Chinese proverb
see no evil hear no evil speak no evil
it’s frequently misused in the West to
imply that you’re avoiding moral
responsibility by looking the other way
but on the contrary it’s a Buddhist
summary of the Confucian tennant look
not at what is contrary to propriety
listen not to what is contrary to
propriety speak not what is contrary to
propriety make no movement which is
contrary to propriety this reminds us to
avoid dwelling on evil thoughts and to
instead delight in creating Concord or
more recently marie curie be less
curious about people and more curious
about ideas we used to have to cope with
yellow journalists tabloid journalists
now we have to cope with yellow
chatterbox tabloid chatter BOTS you used
to read news stories now the news
stories read you so don’t be a
stereotype don’t be a cliche
don’t let artificial gossips make you an
unwitting part of the divisive nasir
artificial gossip try to reorient take
your perspective
when artificial gossip mongers slander
or attack someone try instead to relate
when artificial quit knocks are
listening in on you translate your ideas
into ideas worth spreading as a species
we humans face major survival challenges
climate change vast wealth disparities
arms proliferation our only hope maybe a
eyes we cannot afford to turn them into
artificial gossips take personal
responsibility for being mindful being
aware of AIS when they are watching you
when they’re hearing you when they’re
speaking to you especially in this brave
new world of artificial gossips see no
evil hear no evil speak no evil thank you
what’s universal to humankind people
have claimed all sorts of lofty things
music language reason mastery of fire
and cooking in cuisine or even opposable
thumbs
but sadly enough what’s most universal
of all might be gossips every single
culture in the history of humankind has
been plagued by gossips humans religions
through the ages have proclaimed
stricture after stricture against gossip
Christianity in the New Testament Islam
in the Quran Buddhism in the Pali Canon
it goes on and on which tells you how
bad the problem really is but at least
gossips used to be human I’ve been
watching our societies become
increasingly populated by nonhumans over
the past generation populated by
artificial intelligences from a trickle
of AIS in the pioneering days to mass
immigration into our societies today
it’s a tectonic shift that our
imaginations still struggle to come to
grips with and I’m guilty of having been
a part of that tectonic shift because
back when I was a young PhD student AI
was dominated by rule-based systems
systems based on mathematical logic and
I fought with my advisor for six years
he gave me a tiny little problem he said
look
we’ve built one of the first dialog
systems you know the sort of ancestors
to Siri and so on and we just have this
little problem the user comes up and
says how do I cut the disk space and we
can tell from the syntactic parser that
the verb is cut and the direct object is
disk space but the problem is when we
try to plug it into the next step the
semantic interpreter it’s got to tell
you what is that next step is it cut as
in cut bread so meaning slice or cut the
grass meaning shorten or cut class
meaning skip or cut the cheese there are
so many senses you just have to go and
write a little module that will figure
out which sense of cut that is do that
over the summer so I went away less than
one week I came back to him and I said
there is no way on earth you can solve
this with logic because this is all
shades of gray this is all ambiguity
this depends on the context this is the
sort of thing that we do so well which
is not based on logic it is not based on
reason it is based on our intuitive
abilities to integrate shades of grey
from many sources and to make that our
own subjective decision and that is the
root of human intelligence and it took
six years of screaming fights not me
screaming my advisor screaming and
finally six months before I graduated he
said to me one day after I’d given him
yet another draft of a chapter he said
wow guy this is the most amazing thing
ever I finally understand whether you’ve
been saying all these years and that
made me the happiest young student ever
and by that time I had been recruited to
go with 400 other faculty who came from
UC Berkeley where I was and Stanford and
MIT and Toronto and Oxford and Cambridge
to go found a super ambitious new
research university in the American
style
Hong Kong which is a remarkable story
that I’d love to share with you over a
beer Sunday but when I arrived I found
myself thinking where should I apply
this new paradigm that I’d been fighting
for to make machines learn by themselves
from big data and I decided to make them
learn how to translate between languages
because Hong Kong like many other places
is struggling with a multilingual
multicultural problem and so
specifically I made machines learn how
to translate between English and Chinese
for the first time and this this is how
I met one of the creators of artificial
gossips that has shaken the news sphere
many of you have probably heard about a
company called Cambridge analytical this
company worked to use AI to gather and
spread information on a massive scale so
as to influence the outcome of voting in
numerous countries including the US and
the UK and what eventually emerged is
the news that Cambridge Anika was backed
by a reclusive brilliant hedge fund
billionaire named Robert Mercer but
before he took those Wall Street offers
20-some years earlier that made him a
billionaire Mercer was a computer
scientist at IBM Watson working on human
language technology or natural language
processing and specifically he was
working on getting machines to learn to
translate between English and French the
Association for computational
linguistics is the main International
Organization for natural language
processing and about six years ago they
decided to establish an honour called
the ACL fellows in its history of over
half a century the field had never had
anything like nobel or fields or fellow
honours so to catch up with that they
decided that on this one
historic occasion they would name 17
founding fellows Danford breakthroughs
that led to a eyes like google
translator microsoft translator or yahoo
translate the two founding fellows they
named where mercer and me which is why
I’ve been struggling and thinking a lot
about the consequences of the work that
we do developing artificial intelligence
and machine learning those consequences
are enormous our societies today already
are no longer exclusively human in a few
short years a eyes have become our
secretaries our navigators our dictation
assistants our translators our shoppers
our matchmakers our curators our
journalists and our news Clippers our
house cleaners our chauffeurs our hedge
fund traders our deejays our customer
service representatives and our gossips
all our cultures and religions have
warned through the ages against idle
chatter quarrels conflicts gossip
slander condemnation calumny all idle
chatter and now we’ve gone and created
chatter BOTS artificial gossips no
gossips do two main things they hear and
they speak maybe you know this word
quidnunc this means a person who seeks
to know all the latest news or gossip in
other words a busybody this is the type
of gossip who likes to hear and on the
other hand you have the gossip monger
this is a person who habitually passes
on confidential information or spreads
rumors this is the gossip who likes to
well artificial gossips are
exponentially dangerous regardless of
whether the gossip is true or false when
gossip is false or private artificial
gossip mongers spread fake news and
confidential information like human
gossip mongers – but exponentially more
dangerously Douglas Adams in the
Hitchhiker’s Guide to the galaxy
observed that nothing travels faster
than the speed of light with the
possible exception of bad news which
obeys its own special laws well now we
have artificial gossip mongers to give
gossip even more exponential reach and
speed gossiping fake news and
confidential information are powerful
ways to change society’s views even in
the absence of evidence and it doesn’t
take that much there’s a recent study
from UPenn and the City University of
London and suggest that even if only
roughly a 25% minority of a group starts
to spread an opinion that is enough to
tip the majority to switch opinions and
in this age of botnets artificial
gossips easily account for more than 25%
of our gossips artificial gossips push
us past the tipping point for fake news
to change society’s views and just like
humans artificial gossip mongers
weaponize the power of suggestion just
like human gossips they spread
unevaluated claims that are based on
superficial appearance they promote
hearsay rather than evidence they
promote stereotyping as an old Korean
proverb says words have no wings but
they can fly a thousand miles and even
when gossip is true artificial quidnunc
gain social power just like human quit
knocks – but exponentially more
dangerously
in the psychology of rumors which is a
social psychology classic from 1947
gordon allport and leo postman point out
how gossip serves to create a sense of
cohesion among various groups of people
but to take a divisive stance against
someone else gossip actually becomes a
social control mechanism that gives
power to whoever does it
gossiping releases endorphins and gives
pleasure propagating gossips helps them
escape from an unpleasant routine escape
from negative feelings or stress gossips
become the center of attention for the
folks who are receptive to their rumors
if you think Game of Thrones Varys is
scary our artificial critics are
potentially far more terrifying masters
of whispers Cambridge analytic afore
example was an artificial quit no
Facebook is an artificial quit them from
which gossip can unintentionally leak
just as happens with most human
equipments it’s hard to keep a secret
and in this case leaking from one
artificial quidnunc to others including
Cambridge analytic oh but Siri Alexa
Cortana Google assistant they’re all
artificial quit notes and again it can
be hard to avoid leaking what you
overhear even Alexa recently
accidentally spread a private
conversation that it overheard in a home
whether our AIS are hearing gossip or
speaking gossip whether their artificial
Quinn monks or artificial gossip mongers
we exist already today in a tangled
Society of human and artificial gossips
so here’s the question I want to ask
each of you have you joined the
artificial gossips are you an unwitting
part
of the network of the little birds have
you joined the artificial gossips in
creating divisions rather than bridging
them gossip ostracizes persons or groups
exponential artificial gossip
disruptively ostracizes persons or
groups exponential divisive ‘no smiie
–nz that we face the threat the
existential threat of an extinction
event what are each of us doing about
our own habits to prevent the
elimination of our civilization you may
have heard the ancient Chinese proverb
see no evil hear no evil speak no evil
it’s frequently misused in the West to
imply that you’re avoiding moral
responsibility by looking the other way
but on the contrary it’s a Buddhist
summary of the Confucian tennant look
not at what is contrary to propriety
listen not to what is contrary to
propriety speak not what is contrary to
propriety make no movement which is
contrary to propriety this reminds us to
avoid dwelling on evil thoughts and to
instead delight in creating Concord or
more recently marie curie be less
curious about people and more curious
about ideas we used to have to cope with
yellow journalists tabloid journalists
now we have to cope with yellow
chatterbox tabloid chatter BOTS you used
to read news stories now the news
stories read you so don’t be a
stereotype don’t be a cliche
don’t let artificial gossips make you an
unwitting part of the divisive nasir
artificial gossip try to reorient take
your perspective
when artificial gossip mongers slander
or attack someone try instead to relate
when artificial quit knocks are
listening in on you translate your ideas
into ideas worth spreading as a species
we humans face major survival challenges
climate change vast wealth disparities
arms proliferation our only hope maybe a
eyes we cannot afford to turn them into
artificial gossips take personal
responsibility for being mindful being
aware of AIS when they are watching you
when they’re hearing you when they’re
speaking to you especially in this brave
new world of artificial gossips see no
evil hear no evil speak no evil thank you
Please follow and like us: