►
Description
The nature of work today is undergoing an unprecedented transformation. As the pace of technological innovation continues to accelerate exponentially, the skills required of workers by industry are changing more rapidly than ever. These newly emerging skill requirements are having a parallel disruptive effect on higher education and training programs, which are struggling to adjust to the dynamic needs of both workers and industries.
At the same time, technological innovation also threatens to displace human workers with robots and artificial intelligence on a massive scale, with some studies predicting hundreds of millions of lost jobs world-wide. Such levels of unemployment can create cascading effects – such as exacerbated inequality, increased poverty, and stunted development – that contribute to social instability on a global scale.
Yet technology is also likely the key to mitigating these issues surrounding the future of work. Breakthroughs in big data and machine learning are creating new opportunities to rapidly match the needs of employers with the abilities of workers; human-machine interfaces promise to revolutionize how work and workers are defined; and intelligent machines may lead to drastic improvements in human quality of life. In this session, our panelists will explore some of these important questions related to the future of work, training, education, and technology.
A
B
A
Those
of
you
who
may
not
be
familiar
the
south
east
data
hub
mission,
it's
a
serve
as
a
catalyst
and
strengthen
public/private
partnerships,
so
connecting
academic
researchers,
government
and
nonprofits
to
apply
data
science
to
regional
and
national
challenges.
Before
we
get
started,
I'd
like
to
mention
a
few
things
for
those
of
you
who
tweets,
please
use
the
hashtag
as
bdh
for
South
big
data
hub
18,
so
s,
bdh,
18
and
hashtag.
Bd
hubs
will
reach
all
the
hub
audience
across
all
four
hugs
in
the
United
States.
A
We
also
have
a
Twitter
account
at
South,
big
data
hub,
so
welcome
to
the
city
of
big
data
hub
data
science
roundtable
series,
which
we
host
regularly
it's
a
monthly
series
that
provides
an
open
forum
for
exploring
the
emerging
research
challenges
in
data
science
and
identifying
potential
solutions.
Our
discussion
today
focuses
on
the
future
of
work
and
tellement
machines,
automation
and
societal
impacts.
A
Joining
us
is
our
moderator,
Molly,
kinder
who's.
A
senior
advisor
of
the
feature
work
at
new
America
and
for
distinguished
panelists
Molly
is
also
a
fellow
at
Georgetown.
University
speak
Center
for
society,
social
impact
and
innovation
and
adjunct
faculty
at
Georgetown
support
School
of
Public
Policy,
where
she
is
teaching
a
graduate
seminar
on
the
social,
economic
and
policy
implications
of
AI
and
I'd
love
to
take
that
class.
As
Kevin
Grafton
is
a
distinguished
professor
of
information
science
in
the
School
of
Information
science
at
Syracuse
University,
he
is
currently
serving
as
the
associate
deeper
research.
A
His
research
examines
new
ways
of
organizing
made
possible
by
the
extensive
piece
of
information
and
communications
technology
and
Kevin
and
I
know
each
other
also
from
the
citizen
science.
Research
community,
Gordon
Friedman
is
the
president
of
the
National
Laboratory
Fred
for
education
transfer
transformation
throughout
his
career.
Gordon
has
helped
to
evolve
the
fields
of
education,
technology
and
online
learning
and
is
now
an
advocate
for
broad-based
change
across
the
K
to
career
spectrum.
Dr.
Joanna
hi
Marvin.
A
Thank
you,
I
apologize
for
my
mr.
minaldi.
She
is
the
assistant
professor
at
the
University
of
Pennsylvania
School
of
Social
Policy
and
practice,
and
a
faculty
research
fellow
at
the
National
Bureau
of
Economic
Research.
As
an
economist.
She
studies
the
labor
market
to
craft
policies
that
can
enhance
employment,
productivity
and
Economic
Security
and
to
close
out
our
powerhouse
of
a
panel
today
is
dr.
A
shade
shutters
who
do
research
scientists
for
the
global
security
initiatives
at
Arizona,
State
University
he's
also
a
PI
on
the
West
big
data
hub
so
exciting
to
have
multi
hub
participation
as
research,
funded
primarily
by
the
National
Science
Foundation,
and
the
Department
of
Defense
uses
big
data
to
understand
the
complex
dynamics
of
urban
systems
and
labor
markets.
So
with
that
I'll
hand,
things
over
to
Molly
to
moderate
our
panelists,
we'll
get
started
with
5-minute
opening
remarks,
followed
by
some
moderated
questions
with
Molly
and
then
we'll
open
it
up
to
Q&A.
C
C
This
is
the
the
issue
that
keeps
me
up
at
night
and
actually
gets
me
to
work
in
the
morning.
So
it's
both
something
that
concerns
me,
but
also
is
an
issue
that
I'm
really
motivated
to
be
part
of
it
new
America.
We
have
a
work
workers
and
technology
team
that
I
am
part
of
and
in
in
addition
to
some
of
my
research
that
I
started
at
Georgetown
and
I.
C
Is
it
as
a
major
societal
question
and
it's
one
that's
top
of
mind,
of
journalists,
of
policymakers
and
of
the
public?
Frankly,
so
I'm
sure
many
of
you
have
seen
any
number
of
newspaper
headlines
about
robots
coming
and
usually
the
newspaper
headlines
tend
to
be
fairly
negative
in
nature
and
that
mood
is
reflected
in
the
public
mood.
C
So
pew
did
a
poll
that
came
out
this
past
fall
that
showed
that
about
double
the
amount
of
respondents
in
America
expressed
worried
as
opposed
to
enthusiasm
about
the
role
that
this
technology
is
going
to
play
in
work
and
policymakers
from
mayor's
to
governor's
to
to
Congress
are
starting
to
wrap
their
heads
around
what
this
means
and
what
we
need
to
do
to
prepare
and
hovering
over
all
of
this
interests.
As
this
big
policy
question
and
research
question,
which
is,
is
there
something
different
today
about
the
technology?
That's
impacting
work?
C
Is
this
different
than
previous
waves
of
technological
revolution
and
innovation
that
we've
had
in
the
past?
The
anxiety
is
certainly
not
different.
You
can
see
newspaper
headlines
from
40
years
ago
from
100
years
ago
that
sound
the
same
alarm
bells
that
were
seeing
today,
but
in
previous
waves
of
technology
automation.
There
has
certainly
been
disruption
in
some
displacement,
but
on
net
jobs
have
continued
to
grow
and
prosperity
has
advanced.
C
There
is
a
question
that
everyone's
grappling
with
is:
is
there
something
different
about
the
nature
of
the
technology
that
could
be
more
disruptive
and
more
transformative,
and
potentially
on
that
we'll
see
less
jobs?
Economists
are
pretty
evenly
divided
on
this,
for
lots
of
reasons
and
shade
is
going
to
be
turning
to
some
of
those
questions
in
a
moment.
But
even
if
you
take
the
optimistic
scenario
which
is
on
net,
we
won't
see
a
hemorrhaging
of
jobs.
We'll
still
have
enough
work
to
go
around
still.
C
That
presents
a
pretty
major
challenge
because
there
is
well
there
isn't
consensus
on
the
net
number
of
jobs
we're
going
to
be
seeing.
There
are
some
other
other
bits
of
consensus,
which
is
that
AI
technology
will
change
jobs,
maybe
most
jobs
and
in
some
cases
quite
dramatically.
It's
also
changing
the
skills
and
education
required
to
connect
to
that
opportunity.
C
Biasing
away
from
routine
lower-skilled,
predictable,
work
into
more
complex
cognitive,
creative
and
empathetic
work,
and
even
if
the
sort
of
on
net
work
and
prosperity
is
better
off,
there
will
certainly
be
strong
distributional
impacts,
whether
Geographic,
socio-economic
or
gender,
four
just
to
take
a
few,
so
we're
gonna
dive
into
some
of
these
questions.
First,
what
do
we
know,
and
what
do
we
not
know
about
how
work
is
going
to
how
technology's
gonna
impact
work
and
I'm
especially
excited
that
this
panel
has
really
brilliant
minds
on
who
works?
C
Thinking
about
what
what
do
we
actually
do
about
this?
What
are
some
innovative
solutions
that
we
can
bring
to
the
table
both
from
a
policy
point
of
view,
but
even
from
an
innovation
point
of
view,
as
well,
so
with
that
I'm
gonna
turn
things
over
to
shade
who's,
going
to
walk
us
through
a
little
bit
more
of
this
question
of
what
do
we
know
about
how
work
is
impacting
excuse
me,
technology
is
impacting
work
and
I
just
want
to
reiterate
the
prompt
in
the
beginning.
D
D
One
of
the
places
to
start
is
simply
that
three
million
people
globally
every
week
move
into
cities
and
the
the
primary
reason
for
that
move
is
that
they're
looking
for
employment
opportunities,
so
I
think
we
all
agree
that
globally,
the
requirement
or
the
need
for
productive
jobs
is
extremely
important
and
that
there's
more
and
more
pressure
every
day
to
have
those
jobs
available.
But
at
the
same
time
we
are
facing
this
massive
transition
which
Molly
talked
about.
You
could
see
it
variously.
D
However,
it's
still
a
big
question
in
my
mind,
and
probably
most
people
who
grapple
with
this
as
to
what
that
means.
What
does
that
mean
to
replace
human
cognitive
abilities
with
machines
and
an
artificial
intelligence?
Does
it
enhance
the
ability
of
workers
to
do
their
job?
Does
it
replace
workers
outright
and
you
know,
are
we
going
to
see
this
worst-case
scenario
of
hundreds
of
millions
of
people
roving
the
planet
in
search
of
meaning
and
work,
and
are
we
going
to
have
political
issues
of
around
global
or
I'm
sorry
guaranteed
basic
income
or
certain
remedies?
D
Make
sure
this
is
working
here,
so
there
are
a
lot
of
reports
from
different
agencies
for
profit
companies,
academics
that
project
the
impacts
of
this
coming
transformation.
One
of
the
most
widely
cited
academic
papers
is
by
a
couple
of
authors
at
Oxford,
University,
Frye
and
Osbourne.
The
projected
losses,
according
to
their
estimates,
are
shown
on
the
right-hand
column
for
just
a
couple
selected
occupations.
D
For
instance,
if
you
look
at
the
personal,
a
personal
care
aides
projected
losses
using
the
Frey
and
Osbourne
estimates
are
over
a
half
a
million
jobs
lost,
and
these
are
us
numbers
so
in
the
u.s.
that
we
would
experience
over
a
half
million
jobs
lost,
while
the
Bureau
of
Labor
Statistics
is
still
projecting
increase
of
three-quarters
of
a
million
jobs
in
the
next
ten
years.
C
E
Great,
thank
you
very
much.
It's
also
a
pleasure
for
me
to
be
here
today.
I
I
know
she
did
that
a
lot
of
the
media
attention
around
this
issue
of
work
and
automation
is
really
focusing
on
the
notion
of
widespread
job
loss
and
in
contrast,
what
I
want
to
call
our
attention
to
is
how
it's
going
to
affect
the
work
that
people
are
actually
doing
so
the
day-to-day
hour-to-hour
kind
of
activity.
E
You
know
the
place
where
people
really
spend
most
of
their
time
and
the
perspective
I'm
going
to
take
on
that
is
to
consider
a
job
as
a
collection
of
tasks
that
the
workers
have
to
perform.
Of
course,
there's
a
lot
more
to
work
than
just
doing
tasks,
but
that
perspective
I
think
is
helpful
in
thinking
about
how
the
automation
is
going
to
have
an
impact,
and
so
this
is
actually
not
a
new
way
of
thinking
about
things,
for
example,
and
pulling
up
the
site
here.
E
The
Department
of
Labor
funded
an
effort
to
create
a
database
of
different
kinds
of
jobs,
and
one
of
the
things
that
that
contains
is
a
list
of
the
tasks
that
are
performed
so
I.
Just
pulled
up
a
random
one,
a
cargo
and
freight
agent,
and
you
can
see
at
the
bottom
that
the
database
is
listed
21
tasks.
It's
interesting
to
note,
though,
that
there's
very
little
theorizing
about
these
tasks
in
the
database.
E
It's
just
plain
English
descriptions
and,
in
part,
that's
because
there's
a
great
diversity
of
tasks
that
are
done
so
I
think
one
of
the
questions
is
just
whether
we
can
describe
what
work
is
in
a
somewhat
more
precise
fashion,
but
this
perspective
is
still
useful
in
thinking
about
the
impacts
of
automation,
because
the
tasks
are
different
in
how
automatable
there
are
and
an
answer
to
the
question
of
what's
different.
This
time.
E
I
think
one
thing
we
can
see
is
that
Shane
mentioned
the
the
kinds
of
things
that
can
be
automated,
have
shifted
from
purely
physical,
applying
more
energy
to
sort
of
routine
data
processing.
You'll
notice
that
the
top
one
here
is
just
keep
records,
something
that
machines
are
much
better
at
than
people.
E
So
this
perspective
is
the
basis
for
various
kinds
of
analyses,
the
one
that
was
mentioned
earlier
about
what
percent
of
jobs
could
be
automated
was
done,
based
on
basically
an
examination
of
these
sorts
of
kinds
of
databases,
and
so
it
also
helps
us
to
see
some
of
the
impacts.
So,
first
of
all,
if
we
get
rid
of
the
stuff
on
the
left,
give
it
to
a
machine
that
applies.
Maybe
we
need
fewer
people
to
do
the
smaller
amount
of
work.
That's
left
on
the
right.
E
It
also
has
clear
implications
for
the
kinds
of
skills
that
are
needed.
If
you,
if
you
notice
the
things
on
the
right
might
be
ones
that
do
require
at
a
more
education
or
emotional
intelligence,
more
managerial
ability-
and
it
also
could
be
that
we
start
to
see
redistributions
of
these
tasks.
So
new
kinds
of
organizations
of
the
work
as,
for
example,
the
tasks
get
distributed
differently
across
different
kinds
of
jobs.
E
The
thing
I
want
to
point
out,
though,
is
that
this
shift
of
the
automation
has
impacts
just
that
go
beyond
just
how
much
there
is
and
who
gets
to
do
it.
For
one
thing,
as
has
been
noted
by
lots
of
people,
there
gonna
be
new
kinds
of
jobs
that
are
created
by
the
automation,
so
you
know
developing
systems,
maintaining
them,
programming,
robots
and
so
on.
E
In
fact,
you
can
see
that
in
the
list
of
tasks
that
I
showed
on
the
previous
slide,
there's
a
set
of
tasks
around
monitoring
and
maintaining
this
system.
So,
for
example,
if
you
thought
about
a
freight
forwarding
clerk
using
one
of
these
systems,
making
sure
that
the
tables
of
prices
or
routing
is
kept
up-to-date,
you
know
it's
a
huge
amount
of
work
and
part
of
the
reason
why
the
automation
is
sometimes
to
get
installed
and
there
can,
even
interestingly,
be
upskilling
requirements
as
the
humans
who
are
now
on
the
outskirts
of
this.
E
So
what
that
implies
in
terms
of
research
is
just
a
need
for
understanding,
not
at
the
level
of
you
know,
an
entire
job
classes
or
hundreds
of
thousands
of
workers,
the
kind
of
gross
economic
statistics
that
are
being
collected
or
hypothesized,
but
really
very
detailed
understandings
of
what's
happening,
and
one
advantage
is
that
we
can
actually
already
see
it
happening
in
certain
bellwether
settings
places
where
the
payoffs
for
the
automation
are
higher
and
where
there's
more
money.
So
in
finance,
for
example,
where
better
decision-making
can
really
have
million-dollar
impacts,
there's
been
a
lot
of
investment.
E
So
what
were
the
design
choices
that
people
had
about
these
tasks
in
their
assignment?
What
has
the
impact
of
the
spin
on
a
variety
of
outcomes?
There's
a
lot
of
emphasis.
Looking
at
this
just
from
the
point
of
view
of
efficiency,
but
I
think
we
also
need
to
think
about
worker
learning
well-being.
We
need
to
think
about
these
impacts.
Are
the
group
level
what
happens
to
a
group
of
workers
when
part
of
the
work
is
now
being
done
by
machines?
E
I
think
a
very
interesting
set
of
questions
is
around
the
kind
of
relationships
we
tend
to
think
of
this
very
much
in
the
terms
of
automation
of
just
you
know
replacing,
but
you
could
think
about
it
as
an
apprentice.
You
know
someone
that
you're
training
and
it's
gonna
get
better
over
time
and
aid
and
and
a
symbiotic.
You
know
if
we
really
think
about
the
potential
for
not
just
replacing
humans
but
augmenting
them.
E
You
know
what
kind
of
a
relationship
does
that
look
like,
and
so
in
general,
how
do
we
make
sure
that
we
effectively
design
these
interactions
at
the
frontier
in
order
to
get
the
most
benefit,
not
just
in
terms
of
economic
efficiency,
but
also
in
terms
of
worker
abilities?
So
that
would
be
my
call
for
action
and
with
that
it
was
optional,
is
be.
C
F
F
It
would
be
hard
to
know
which
part
of
them
is
the
machine
and
which
part
of
them
is
the
human
I
mean
we're
entering
a
culture
where
these
things
are
mixed
and
the
nonprofits
that
I,
founded
and
run
the
National
Lab
for
education
transformation
is
designed
really
to
bring
workforce,
training
and
education
into
the
world
that
we
are
already
in
and
how
we
compute
socially.
So
our
whole
effort
is
to
look
at
what's
necessary.
F
So
if
you
think
about
this
issue
of
automation,
one
of
the
key
points
is
humans,
and
machines
and
algorithms
are
all
going
to
have
to
live
together,
and
the
question
is
well.
How
would
we
do
that?
So
right
now
we
have
some
very
fundamental
problems
in
the
job
markets
in
the
education
markets,
and
that
is
when
you
attend
college
or
a
training
program,
there's
very
little
data
on
what
you've
done
or
the
labor
code
standards
or
any
of
the
other
things.
F
That
might
say
what
you've
done,
what
courses
are
training
you've
consumed
and
then,
where
you
end
up
in
the
labor
market,
so
right
now,
when
you
go
shopping
online
or
you
go
to
expedient
order,
something
you're,
all
this
coke
processing
is
happening.
So
the
work
that
we're
engaged
in
is
to
say
what
are
the
underlying
data
standards?
Look
like
how
do
we
move
data
from
one
place
to
another?
How
can
effectively
we
move
from
a
binary
situation
where
robots
might
be
doing
a
lot
of
work,
we're
not
actually
communicating
with
them.
F
The
data
doesn't
flow
back
and
forth,
so
you
could
imagine
and
workflows
in
the
future
where
we're
intimately
involved
with
different
machine
processes.
That's
I
have
to
run
on
data
standard.
So
there
are
a
number
of
efforts
nationally
to
look
at
how
we
define
skills
in
how
we
get
those
in
the
data
and
metadata,
but
the
transport,
the
transport
data,
how
we
move
this
around
is
not
really
it's
not
really
at
a
very
high
level.
The
other
thing
I
want
to
think
about
is
again:
it's
not
us
in
them.
We've
created
them.
F
So
the
question
is:
are
there
robots,
especially
with
people
as
Schade
mentioned,
whose
jobs
are
at
risk
or
the
opportunity
youth
in
the
16
to
24
category
are
their
robots
that
can
actually
help
them
find
work,
so
in
other
words,
if
you're
a
17
year
old
without
work
and
not
in
school?
Where
do
you
go
today
and
I
think
anybody
on
this
call
widely?
We
have
a
hard
time
saying:
well,
where
does
that
person
go
to
find
a
job?
Where
do
they
go
to
find
the
training
they
need
for
a
job?
F
So
I
don't
think
we're
deploying
robots
right
now
to
assist
people
in
finding
work
and
then
that
work
maybe
work
where
they're
working
with
a
robot
right
or
they're
they're
co-constructing,
something
so
basically
our
work
is
to
say
what's
the
data
infrastructure.
All
of
this
is
going
to
run
on
what
are
the
standards
and
were
involved
in
a
number
of
efforts
to
get
people
to
look
at
those
issues
so
that
these
conversations
become
much
more
realistic
in
terms
of
what
can
be
done.
F
C
Great
thanks,
green,
that's
really
interesting
and
you
wanna
I'm
excited
to
bring
you
into
the
conversation.
As
we
now
shift.
Gordon
did
a
nice
job
of
teeing
up
some
potential
solutions.
So
Gordon
was
talking
about
how
do
we
leverage
data
and
robots
to
actually
help
people
connect
with
opportunity?
A
lot
of
the
anticipating
that
we
may
be
facing
that
individual
workers
are
gonna
face
disruption,
whether
outright
displacement
or
periods
of
unemployment
or
underemployment,
as
they're
switching
occupations
or
skills.
C
There's
been
a
lot
of
interest
in
sort
of
safety
nets
and
how
do
we
think
about
supporting
workers
through
different
transitions?
One
of
the
one
that
gets
a
lot
of
attention
is
universal
basic
income,
so
I'm
delighted
to
bring
you
into
the
conversation
to
speak
a
little
bit
about
your
research,
which
looks
both
at
shed
some
new
insights
onto
some
of
the
debate.
It's
about
universal
basic
income,
but
also
looks
at
various
policy
levers
for
supporting
Economic
Security.
G
It's
the
idea
of
giving
everyone
in
the
territory
cash
with
no
strings
attached
with
no
conditions,
so
everyone
gets
this
transfer
as
such.
This
is
a
transfer,
therefore,
that
provides-
and
always
there
bed
of
Economic
Security
to
everyone,
and
so
that's
in
large
part.
The
reason
why
this
idea
is
appealing.
F
G
We
are
facing
so
much
uncertainty
from
technological
change
and
how
it
might
affect
affect
jobs
and
workers
livelihoods.
So
my
own
research
looks
at
some
of
the
impacts
of
basic
income
style
policies
that
have
been
done
in
the
past
and
so
before.
I
talk
about
my
own
most
recent
research
I
just
want
to
give
our
listeners
an
overview
of
what's
been
found
before
this
idea
is
not
new.
It's
been
around
for
at
least
a
couple
of
centuries
and
in
fact,
the
scientists,
the
government,
the
u.s.
G
What's
the
point,
so
you
know,
these
studies
have
looked
at
that
the
impact
on
work
and
perhaps
surprisingly,
they
found
oftentimes
no
impact
at
all
and
sometimes
a
small.
You
know
reduction
in
hours,
but
not
very
much
so
the
impact
range
from
zero
to
a
small
reduction
in
hours
and
that's
very
interesting
to
know
for
us
in
case
we're
worried
about
some
of
the
downfalls
of
this
policy.
Now.
G
And
they
have
found
important
positive
impacts
on
education
and
health,
especially
for
the
most
disadvantaged
youth.
So
that's
the
picture
of
what's
been
done
before
and
that's
where
my
new
work
on
Alaska
comes
in.
So
it
turns
out
that
Alaska,
which
is
the
u.s.
state,
has
transfer
scheme
like
a
basic
income.
It's
not
called
the
basic
income.
Is
the
Permanent
Fund
dividend,
so
it
comes
from
Alaskan
money
that
was
put
into
this
fund
from
oil
and
minerals.
They
have
a
sovereign
fund
which
is
basically
a
fund.
That's
invested
in
a
diversified
portfolio.
G
The
dividend
from
this
fund
comes
every
year
to
every
Alaskan.
Anyone
who
has
lived
in
Alaska
for
a
year.
Anyone
who's
illegal,
including
refugees,
is
a
title
to
to
this
transfer,
so
we'll
be
looking
at
my
co-author,
Damon
Jones
and
inverse
from
Chicago
and
the
impact
of
this
on
work
in
Alaska,
and
the
reason
why
this
is
new
and
different
from
what's
been
done
before
is
because
this
is
a
transfer
that
everyone
in
Alaska
gets
well.
G
Prior
studies
were
small-scale
experiments,
just
a
few
people,
so
here
we're
able
to
look
at
the
whole
of
Alaska
and
compare
it
with
similar
states
and
see
if,
after
Alaskans
start
receiving
this
transfer,
they
stopped
working
or
what
and
so
basically
what
we
find
is
that
this
transistor
going
on
for
thirty
years.
So
it's
really
you
know.
If
something
was
going
to
happen,
it
would
have
happened.
G
Basically,
we
find
no
effect
Alaskans,
although
less
likely
to
be
employed
that
residents
of
other
similar
states,
but
they
did
increase
their
part-time
work
and
so
consistent
with
some
of
the
prior
research.
There
is
some
reduction
in
hours
for
some
people,
but
in
terms
of
overall
employment,
we
find
a
zero
effect
and
then
we
go
on
to
argue
that,
like
the
zero
fact
is
due
to
the
fact
that
when
you
give
people
money,
it
has
an
impact
on
the
economy,
which
is
a
positive
impact.
G
So
if
you
think
about
local
restaurants,
for
example,
as
people
get
a
lot
of
cash,
some
of
this
cash
goes
into
spending
locally
in
restaurants.
So
this
restaurants
now
need
to
hire
more
cooks
and
waiters
in
order
to
satisfy
this
demand,
and
this
effect
is
positive
and
can
countervail
any
negative
effect
from
just
you
know,
giving
cash
to
people.
So
we
believe
that
this
cash
stimulates
the
economy
and
therefore
the
overall
effect
on
work
is
more
positive
than
you
might
have
thought
if
you
were
just
looking
at
every
individual.
G
So
in
conclusion,
you
know
I
think,
there's
much
more
research
to
be
done,
but
certainly
the
effects
of
basic
income.
Stat
programs
that
have
happened
before
are
mostly
positive,
and
certainly
the
effects
on
work
again
are
zero
to
small
and,
as
such,
I
feel.
This
is
a
policy
that's
worth
thinking
more
about
in
our
arsenal
of
possible
ways
of
dealing
with.
G
C
Thanks
you
wanna
it's
it's
it's
encouraging
to
hear
that
your
research
is
indicating
that
this
might
have
a
positive
impact.
Staying
on
this
theme
of
solutions
since
I
think
there's
so
much
fear
about
this
issue,
it's
it's
helpful
to
hear
big
ideas
for
what
can
be
done.
Gordon
I
wanted
to
go
back
to
you.
You
you're
in
your
work.
You've
done
a
lot
of
rethinking
education
and
working
with
education
system.
What's
your
review
of
how
education
and
training
needs
to
be
reimagined,
given
some
of
the
the
issues
that
both
Kevin
and
shade
have
raised?
Well,.
F
You
know
in
the
United
States
we
have
institutions
and
employers
very
different
than
what
you
would
see
in
Europe,
where
you
have
a
real
human
capital
model
and
the
human
capital
model
would
say
that
education,
training
and
work
in
personal
development,
social
interaction,
they're
all
part
of
the
same
fabric.
The
problem
that
we
have
in
the
United
States
is
that
we
don't
have
this
point
of
view.
Even
though
human
capital
Theory
came
out
of
the
United
States.
F
So
what
what
has
to
happen
is
that
you
can
have
a
situation
in
k-12
as
an
example,
we
don't
do
big
data
analyses
in
in
k-12,
so
we're
looking
at
comparative
data
across
k12.
We
know
from
the
work
of
Raj
Chetty
and
others
that
the
third
graders,
if
you're
not
doing
well
in
third
grade
in
math
and
English,
you
can
look
at
the
trajectory
forward.
So
why
aren't
we
doing
something
about
that?
Why
isn't
that
family
in
some
kind
of
a
system?
F
Now,
just
yesterday,
there's
a
San
Diego
School,
District,
that
by
the
time
you're
in
eighth
grade,
you
will
have
been
exposed
to
54
different
careers.
Now,
if
they
move
that
student
into
what
I
would
call
a
universal
transcript,
so
their
exploration
of
what
they
do
in
school
and
the
standards
what
they
do
outside
that
becomes
a
form
of
data,
then
you
look
at
when
that
student
gets
ready
to
explore
worker
College
again
by
the
data
we
have
to
have
an
ability
to
give
them
a
view
of
what's
possible.
F
What
to
explore
one
of
the
largest
problems
we
face
and
I
think
this
is
the
the
energy
the
machines
have.
Is
that
most
people
don't
know
what
the
skill
based
jobs
are
today.
So
they're
plenty
there's
there
plenty
of
open
jobs
that
pay
up
to
$100,000.
Where
you,
you
may
not
need
a
high
school
college,
a
BA,
but
you
might
need
some
training.
Well,
most
people
don't
know
what
these
jobs
are.
So
one
of
the
other
things
we're
looking
at
is:
how
does
how
can
we
have
media?
F
F
We
have
plenty
of
economic
outcome
data
looking
at
shade
pointed
out
in
people's
educational
levels,
but
now
we're
going
to
have
more
data
on
their
training
and
so
forth,
but
at
a
community
college,
where
their
occupational
programs
we're
lining
up
people
with
work,
but
we're
not
using
data
systems
to
do
it.
So
again,
we're
looking
at
how
you
get
into
the
the
data
of
courses
and
in
occupational
programs
in
looking
at
their
outcomes
getting
those
again
in
some
kind
of
database.
F
B
F
We
have
a
problem-
that's
that's!
That's
international
and
it's
exacerbated
in
the
United
States
that
we
have
a
high
number
of
16
to
24
year
olds,
who
are
not
in
education
and
they're,
not
in
the
workforce,
reaching
them
giving
them
a
voice.
So
part
of
our
work
is
to
think
about
how
social
media
vehicles
with
data
behind
them
and
media
can
start
playing
a
role.
So
back
to
my
original
point,
we
haven't
made
a
culture
for
this.
We
certainly
have
made
a
social
interaction
culture
on
the
web.
F
C
Thanks
Gordon
I
appreciate
it.
Kevin
I
was
really
struck
in
your
excellent
presentation
that
a
clear
takeaway
it
seems
of
going
beyond
just
the
headlines
of
job
losses
or
looking
at
the
overall
risk
of
a
specific
occupation,
but
taking
it
down
a
step
in
to
looking
at
the
specific
tests,
some
of
which
might
be
displaced
by
technology,
but
others
the
task
nature
shifts.
What
do
you
think
that
policy
implications
of
that
are?
So
what
is
that,
in
terms
of
the
trends
you're,
seeing
in
the
kind
of
skill
shift?
What
is
that?
C
What
does
that
say
about
both
the
kind
of
foundational
skills
that
we
need
to
make
sure
all
workers
are
getting,
but
then
the
specific
skills
that
related
to
that
job?
So
you
mentioned,
for
instance,
emotional,
interpersonal
skills
being
more
of
a
foundational
skill,
but
then
there
might
be
some
specific
skills
of
learning
to
work
with
the
specific
technology.
That's
coming
to
play
so
I'm
curious,
taking
a
step
back
from
your
presentation.
You
know
what
what
does
that
imply
from
from
how
we
need
to
better
prepare
workers.
E
It's
interesting
question
and
I
think
a
really
important
one
I'm,
not
I'm,
not
really
that
much
of
a
policy
person,
and
so
so
it's
a
little
difficult
for
me
to
answer
I.
Think
one
clear
implication,
as
you
said,
is
that
there
are
certain
sets
of
skills
that
are
likely
to
be
less
automatable
in
the
sort
of
near
and
medium
term
and
so
making
sure
that
people
are
able,
with
those
I
think
is
important
and
I.
E
You
mentioned
emotional
intelligence,
for
example,
you
know,
there's
a
variation
in
the
population
and
and
if
that
turns
out
to
be
a
key
skill,
then
people
who
are
lower
in
that
would
would
obviously
be
disadvantaged
and-
and
it's
actually
an
interesting
question-
whether
that's
something
that
can
be
trained
or
not,
but
it
may
be
that
understanding
the
extent
to
which
you
can
be
trained
and
making
sure
that
that's
built
into
educational
systems
is
important.
The
other
thing,
which
is
very
clear,
is
the
ability
to
work
with
the
system
and
understand.
E
A
lot
of
that
would
have
been
on
the
job
right
that
you
would
have
started
it
off
as
an
entry-level
person,
doing
work
and
moved
your
way
up
and,
as
a
result,
personally
experienced
a
lot
of
those
things
and,
to
the
extent
that
that
kind
of
entry-level
assistance
sort
of
stuff
is
actually
automated.
It's
I
think
very
difficult
because
you
don't
have
that
pipeline
in
you
know
you
need
to
come
in
with
a
certain
level
of
expertise,
and
so
I
think
that
that
also
has
some
implications.
E
C
C
Just
this
past
month
came
out
with
a
big
new
report
on
automation
and
the
OECD
one
of
their
key
policy
findings
is
that
yeah
there's
a
u-shape
risk
and
that
young
people
under
25
are
particularly
vulnerable,
because
so
many
of
those
jobs
that
don't
require
any
prior
training
like
I,
was
a
category
like
men's.
When
I
was
in
high
school,
you
know
you
just
get
trained
on
the
job.
Those
are
that
those
tend
to
be
some
of
the
jobs
most
vulnerable
to
computerization
and
automation.
C
E
E
A
E
Drawing
diagrams
and
every
time
you
needed
to
change,
someone
would
have
to
redraw
it,
and
so
it's
really
boring
tedious
work,
but
it
was
the
starting
position
for
a
lot
of
architects.
Now
the
people
have
CAD
systems
that
entire.
You
know,
the
roomful
of
people
doing
drafting
have
disappeared,
but
those
were
the
entry-level
jobs
that
got
you
into
the
PD
firm
and
got
you
trained
up
in
the
actual
practice
to
being
an
architect.
E
C
Great
I've
heard
similar
examples
with
law
with
folks
who
sort
of
had
entry-level.
You
know,
search
positions
that
are
being
automated,
so
it's
it
any
question.
I'm
gonna
be
turning
momentarily
to
the
audience.
We
have
at
least
one
question.
That's
come,
but
before
I
do
that
I
just
want
to
quickly
go
back
to
you
wanna.
You
have
some
additional
work
beyond
your
work.
Looking
at
universal
basic
income,
that's
looking
at
other
ways
of
providing
economic
security,
including
through
unemployment.
G
G
C
Great
thanks
you
so
I'm
gonna
turn
to
some
questions
that
Eric
has
has
posed.
She's
got
it's
in
two
parts.
Let
me
read
this
and
then
I'll
turn
to
the
panelist.
Is
it
a
zero-sum
game,
with
redistribution
of
work
controlled
by
training?
Past
transformations
have
left
a
large
group
of
people
demoted
from
the
middle
class
to
low-income
class
and
often
poverty.
Eric
Ries
is
a
really
good
point.
C
So
I'm
not
sure
some
of
you
have
followed
David
autor
from
mi
t--'s
work
he's
looked
a
lot
at
previous
pasts,
a
30-plus
years
of
technological
automation
and
has
found
a
polarization
of
wages
because
of
Technology.
So
a
lot
of
formerly
middle-class
jobs,
clerical
jobs
replaced
by
email
and
you
know
Microsoft
Word
type
applications
manufacturing,
which
was
a
middle
middle
wage
job
that
tech,
not
that
it
was
the
middle
skilled
jobs
that
were
worse,
most
impacted
by
technology.
C
Some
of
those
folks
acquired
more
skills
and
moved
up
into
higher
paying
jobs,
and
quite
a
few
of
them
dropped
down
to
more
low
paid
service
jobs.
So
there
is
a
big
question
about
the
current
wave
of
automation.
Is
this
one
that
is
similarly
going
to
impact
middle
skilled
workers
and
polarizing
in
the
ways
where
the
rich
get
wealthier
and
a
lot
of
people
fall
down
to
poverty
or
actually
I
think
going
to
shade
slides
about?
C
Actually
the
lowest
wage
workers
are
the
most
vulnerable,
so
we
would
see
income
inequality
potentially
growing,
but
not
necessarily
the
same
impact
of
displacing
the
middle
so
shade.
Why
don't
I
turn
this
over
to
you
I'm
curious
about
your
perspectives
of
whether
this
this
type
of
automation
follows
previous
waves
and
what
is
that
going
to
say
about
whether
this
is
zero-sum
and
how
it's
going
to
impact
inequality?
C
D
And
I
I
am
familiar
with
the
the
work
on
the
past
transitions
and
everything
that
I've
seen
indicates
that
this
one
is
more
towards
the
low
end
of
the
wage
spectrum.
Unlike
previous
ones,
and
one
thing
that
worries
me
is
a
recent
Pricewaterhouse
report
estimated
that
automation
was
going
to
have
the
benefit
in
the
next
10
years
of
increasing
global
GDP
by
almost
sixteen
trillion
dollars
per
year.
D
But
the
concern
is
that
massive
bump
in
GDP
will
be
shared
by
an
ever
shrinking
number
of
workers,
if
you
believe
the
estimates
of
the
automation
displacement
and
so
what
what
you'll
see
globally
is
a
really
rapid
acceleration
of
wage
inequality,
and
that
has
repercussions
far
beyond
just
the
economic
realm.
The
I
think
Leah
said
earlier.
D
We're
researching
these
topics
at
Arizona
State
right
now
about
how
effective
global
basic
guaranteed
basic
income
would
be
in
mitigating
those
effects.
And
one
of
the
interesting
findings
is
that
for
people
that
were
displaced
involuntarily,
sometimes
that
income
is
still
not
enough
to
replace
the
meaning
in
their
lives
that
work
provided
beyond
just
income
and
so
I
think
as
policymakers
struggle
with
how
to
deal
with
people
that
are
displaced.
There
has
to
be
a
discussion
beyond
just
replacing
their
lost
wages,
but
something
that's
a
little
more
difficult,
replacing
this
meaning
in
their
lives.
C
E
I
think
it
would
be
I
think
it's
important
to
separate
the
impacts
on
the
skill
level,
with
the
impact
on
income.
I.
Think
as
Yanis
research
points
out,
the
questions
about
income
are
not
technologically
driven
there
choices
that
we
make
either
managerially
within
organizations,
whereas
say
as
a
society,
and
so
you
know,
the
increase
in
polarization
of
salaries
is
already
happening.
E
It's
been
happening
for
a
while
I,
don't
think
the
complete
driver
now
is
technologically
it's
it's
policy
decisions
that
are
being
made
about
letting
people
with
lots
of
income
set
rules
that
allow
them
to
concentrate
even
more
the
wealth,
and
so
these
questions
about
wealth,
I,
think
and
wealth
disparities
need
to
be
separated
from
skilled
questions.
The
skill
questions
are
quite
important.
You
know
if
you
need
an
increasing
skill
just
to
get
a
job.
C
So,
if
there's
a
pent
up
demand
for
lots
of
very
skilled
labor
and
not
enough
supply,
that
will
drive
income
increases
and
on
the
flip
side,
if
what
shade
is
saying
is
correct,
that
some
of
the
lowest
skilled
and
lowest
wage
workers
are
most
vulnerable
and
automation
can
be
impacting
a
lot
of
occupations
at
the
same
time,
I
think
that
does
not
speak
well
for
for
the
the
markets
delivering
higher
wages,
but
I
agree
that
there's
clearly
political
ways
to
intervene.
So
we
have
yeah.
A
G
G
C
You
might
think
that's
right.
I
also
think
there's
different
time
likes,
I
think
for
some
of
the
higher
skill
jobs
to
be
automated.
The
technologies
gonna
have
to
get
a
lot
better
than
it
is
today,
so
the
MIT
folks
are
techno
optimists
that
will
get
there
okay.
So
let
me
turn
to
some
of
these
great
questions
that
we're
getting
I.
Think
it's
Megan,
sorry
aren't!
Your
name
has
been
cut
off
in
mine.
Can
you
share
what
you
think
are
some
of
the
most
pressing
research
questions
at
the
intersection
of
your
disciplines?
C
Where
do
you
see
the
most
transformative
opportunities
in
the
research
landscape?
I
like
this
notion
of
intersection,
so
not
in
sort
of
one
discipline,
but
where,
where
some
of
this
sort
of
transformative
opportunities
that
might
cross
disciplinary
boundaries,
I'm
not
sure
if
anyone
has
Gordon
come
in
anyone
you
want
to
let.
F
Me,
let
me
jump
in
our
our
work
is
primarily
building
cross
sector
teams
to
look
at
these
problems
and
I
think
that
if
you
look
at
the
work
of
the
OECD,
which
is
great,
even
though
we're
the
largest
contributor
to
them,
we
are
the
least
of
the
consumers
in
so
I
think
we
don't
have
this
dialogue
cast
at
the
right
level
and
so
for
the
research
intersections
we
have
in
the
United
States.
We
have
a
lot
of
organizations
that
compete
and
kind
of
jockey
to
be
oh
I'm,
the
guy
who's
going
to
figure
out
training.
F
Our
organization
knows
this
better
than
somebody
else,
so
I
think
we
need,
on
the
intersection
side
to
have
a
sort
of
larger
scale
collaboration.
So
what
we're
doing
today
is
just
part
of
a
larger
national
conversation
where
we
organize
it.
So
I
think
there's
tremendous
intersections
and
again
I'm
going
to
go
back
to
data
we've
gotten
a
lot
of
data,
but
we
don't
have
organizations
that
are
bringing
it
together
pointed
at
these
questions
in
a
way
that
can
be
tested
and
scaled.
F
E
I'm
gonna
put
up
the
slide
so
with
some
colleagues,
we
recently
got
funding
from
National
Science
Foundation
to
run
a
research
coordination
network
which
we're
calling
work
in
the
age
of
intelligent
machines.
So
RCN
is
an
interesting
funding
mechanism
because
it
does
not
actually
pay
to
do
research,
so
it
doesn't
really
answer
the
call
from
Gordon,
but
it
does
pay
to
bring
people
together
and
so
we're
trying
to
build
a
network
of
cross-disciplinary
researchers
are
interested
in
this
and
and
hopefully
by
bringing
people
together,
create
some
of
those
multidisciplinary
research
projects.
E
So
we
kept
money
to
pay
for
workshops,
we're
going
to
have
a
conference
coming
up
the
first
one
18th
and
19th
of
May
at
Michigan
State,
and
we're
also
hoping
that
this
will
be
a
vehicle
for
sharing
resources.
So
I
was
very
interested
to
hear,
for
example,
that
you
have
a
course
I
think
sharing
syllabi
and
getting
other
people
to
be
able
to
teach
similar
courses
would
be
great
so
that
the
network,
the
URL,
is
Wayne
networked
and
so
I
just
wanted
to
put
that
up
there
to
invite.
C
C
The
difficult
perspective
is
often
within
the
default
perspectives,
often
Healthcare
and
IT,
but
as
we
know,
displaced
workers
may
not
alternative
these
sectors
so
and
I
would
just
want
to
concede
that
most
of
the
energy
in
the
predictions
around
automation
is
what
jobs
are
at
risk,
not
necessarily
the
ones
that
are
on
the
rise
and
that's
the
challenge
I
face
in
my
own
research.
So
shade
or
others
does
anyone
have
thoughts
about
the
more
positive
side,
whether
it's
chicag
directly
created
by
technology
or
that
might
be
created
across
economy
as
productivity
turns
into
higher
wages?
C
D
I
think
the
question
asker
got
the
main
two
sectors
that
are
enhanced
by
technological
needs
of
the
workers,
our
skills
of
the
workers.
On
the
other
hand,
there's
another
sector
that
I've
heard
a
lot
of
people
read
a
lot
that
should
be
enhanced
in
terms
of
increased
employment,
and
that
is
any
occupation
that
requiring
interpersonal
skills,
personal
touch
and
partly
that
is
driven
by
demand,
but
also
by
as
Kevin
put
at
certain
skills.
That's
it
tend
not
to
be
automatable,
so
anything
that
requires
face-to-face
communication.
Interpersonal.
E
And
then
you
know
intervening
when
some
kind
of
problem
arises
and
so
so
that
kind
of
shift
from
people
actually
doing
things
to
people,
monitoring
the
systems
doing
things
and
then
stepping
in
when
they,
when
they
don't
do
it
quite
right.
I
think
you
might
see
that
pattern
repeat
across
a
number
of
different
sectors.
That's.
C
Great
things
come
in
Darryl
had
a
great
great
question
on.
They
challenge
are
the
different
timescales
for
the
problems,
so
the
rate
at
which
jobs
are
lost,
versus
a
length
of
training
or
on-ramp
couriers.
What
tensions
arise
and
how
do
we
think
about
some
of
that
Gordon
I'm,
not
sure
if,
if
some
of
your
education
work
might
be
helpful
and
answering
that
I
think
you're
on
you.
F
We're
deeply
involved
with
the
Workforce
Development
Community
Colleges,
and
they
typically
get
a
couple
breakfasts
a
year
with
employers
to
decide
what
AM,
what
occupational
programs
to
put
together.
Then
it
takes
12
to
18
months
to
make
that
program
at
which
time
it's
very
possible.
The
jobs
don't
exist
anymore,
so
we've
worked
and
we
are
working.
F
We
invite
anybody
to
help
us
on
a
project
called
job
signals
to
try
to
get
employers
to
almost
like
tweet
little
bits
of
pieces
of
information
about
demand
formation,
because
the
when
a
by
the
time,
something
is
in
a
job
posting
and
it's
gone
through
HR
a
it
has
very
little
relevant
information
in
it
and
B
it's
way
too
late.
So
we
need
to
know
regionally
because
reality
is
regional.
What
employers
are
thinking
about.
D
I
would
I
I
just
want
to
add
that
since
I
started
working
on
this
topic,
probably
the
the
most
eye-opening
thing
that
I
have
discovered.
It
has
been
from
working
with
industry
partners,
sort
of
canvassing
them
some
focus
groups
and
and
coming
to
the
realization
that,
in
parallel
to
this
transformation
of
workers
and
automation,
there
will
be,
or
there
will
need
to
be
a
revolution.
Let's
say
in
education
and
the
general
consensus
that
was
sort
of
shocking
to
me
from
industry.
D
They
they're
continuously
updating
every
couple
years
their
list
of
skills
and
that
that
what
they
would
want
to
see
in
the
future
on
a
person's
resume
were
a
list
of
credentials
and
not
degrees,
and
so
to
the
to
the
questioners
question
this.
There
needs
to
be
a
completely
different
time.
Scale
applied
to
the
acquisition
of
skills,
I
believe.
C
Juu
you've
done
a
great
job
of
teeing
up
our
next
question,
which
is
how
is
academia
helping
students
today
ensure
they
have
the
skills
needed,
or
maybe
we
should
say
how
they're
not
helping
a
good
example
of
a
leader
in
this
as
Singapore,
which
is
a
often
cited
example.
Singapore
is
really
embraced
technology
and
automation,
but
made
a
commitment
to
do
exactly
this
kind
of
lifelong
learning.
C
Upskilling
there's
everyone
over
a
certain
age
gets
access
to
funding
things
$500
for
classes,
the
national
university
pivoted
to
create
more
of
these
shorter
skill,
related
courses
that
peak
that
working
adults
can
go
back
and
learn.
So
that's
an
example:
I
would
just
route
of
a
country,
that's
taking
that
seriously,
but
I'm,
not
sure
of
other
panelists.
This
question
of
how
academia
is
or
is
not
shifting?
What
do
we
need
to
do?
Gordon,
please
jump
in
so.
F
F
G
Skills
and
getting
those
two
rights
is,
you
know,
is
no
easy
matter,
because
we
can
use
better
and
more
information
of
this.
Now
that
I
think
the
technology-
and
you
know,
model
and
I
have
worked
on.
This
is
more
and
more
available
for
us
to
do
much
better
job
and
that,
while
doing
that,
we
also
have
to
keep
in
mind
that
you
know,
as
others
have
pointed
out,
demand
is
rapidly
changing,
and
so
we
also
have
to
make
sure
to
film
for
students
bedrock
of
solid.
E
C
C
F
I,
don't
want
to
keep
keep
reacting,
but
again
there,
the
exploration
of
getting
some
of
this
more
into
social
media,
so
that
people
are
communicating
with
each
other.
It
is
essential
and
we're
looking
at
that
issue
and
I
think
it
can't
it's
gonna
have
to
cross
over
into
something
that's
more
in
the
consumer
space,
so
in
other
words,
we
don't
have
on
LinkedIn
right
now
that
a
lot
of
people
on
some
of
these
jobs
would
be
we're
becoming
it
in
that
truck
driver
by
the
way,
if
they've
been
professionally
trained.
F
There
are
all
kinds
of
skills
that
can
be
unpacked
there,
so
so
so
the
necessity
of
doing
this
there's
the
notion
of
a
universal
transcript
that
would
be
able
to
have
your
training
your
skills,
your
your
academic,
certifications,
I,
think
something
like
that
is
going
to
become
very
useful,
especially
again
if
it
can
be
shared,
and
then
people
can
communicate
with
that
person.
Great.
C
I'm
gonna
do
a
little
bit
of
a
rapid
fire
because
I
want
to
get
through
some
of
these
Ranade
ask
Gordon
and
shade:
are
you
working
with
any
technical
or
community
colleges
to
provide
skill
based
IAI
to
help
with
worker
placement?
If
not,
you
know
groups
who
are-
and
let
me
just
keep
going
with
a
few
more
so
that
we
can
respond.
How
do
you
legitimize
the
lifelong
skills
in
the
US?
The
major
universities
have
also
had
500
years
to
build
a
reputation.
What
about
some
of
these
newer
providers
and
last?
C
There
was
a
good
one
up
here,
a
question
for
Kevin
what
are
the
most
exciting
research
questions
and
computer
science
regarding
future
of
work,
so
I've
just
thrown
a
bunch
out
there
in
our
remaining
four
minutes.
If
you
can
be
pithy
and
try
to
respond
to
sound,
that
would
be
great
I'll,
actually,
even
once
as
I
think
we
need
some
follow-up
remarks,
let's
be
like
lightning,
fast,
lightning,.
E
B
F
D
I'll,
just
throw
out
an
example
that
answers
two
or
three
of
these
questions
that
Arizona
State
were
experimenting
with,
including
AI
in
the
delivery
of
education
itself,
we're
slowly
making
all
of
our
curriculum
available
through
online
courses,
but
as
a
student
takes
that
through
an
online
platform,
there's
an
AI
engine
in
the
background
that
is
customizing
it
as
they
go
through
the
process.
So
it's
almost
individualized
for
every
person
that
takes
it
and
thinking
of
that
truck
driver.
We
could
extend
that
model
to
take
them
through
not
just
educational
opportunities
tailored
to
them.
A
Okay,
thank
you.
I
think
this
will
be
the
first
of
several
series
of
roundtables
on
the
future
of
work.
I
hope
to
be
collaborating
with
all
the
panelists
on
this,
including
discussions
about
putting
out
a
white
paper
on
the
on
emerging
research
issues.
What
a
thank
all
the
panelists
and
the
moderator
for
an
outstanding
job
and
thank
you
audience
for
participating.
We
do
have
a
few
announcements
for
the
South
big
data
hub.
A
The
first
is
Microsoft
Azure
training
at
the
Renaissance
computing
Institute
in
Chapel
Hill
on
Tuesday
May,
2nd,
that
is
part
of
the
SAP
hub's
partnership
with
Microsoft
Research,
and
it
is
pre-training
full-day
workshop
offers
a
chance
to
learn
about
the
fundamentals
of
azure,
for
research
training
and
to
talk
one-on-one
with
desert
experts
of
software,
the
systems,
architects,
etc.
About
your
specific
project
or
a
use
case.
Please
check
out
the
website
bit
ly
at
our
slash.
A
Is
your
20
for
more
information
and
we
do
have
some
little
bit
limited
travel
scholarships
to
attend
next,
the
South
Big
Data
innovation
hub
round
table
next
month
will
be
held
on
Thursday,
May
24th
from
noon
to
1:15
p.m.
Eastern,
and
it
will
focus
on
data
science,
recoupment
retention
and
diversity
to
register
in
advance
for
annex
ground
table
or
to
view
the
recording
of
previous
round
tables.
So
to
answer
some
people's
questions.
Yes,
this
this
round
table
has
been
videotaped
and
will
be
archived
and
available
on
our
YouTube
channel.
A
Please
visit
Italy
slash
forward,
slash
south
hub
round
tables.
We
have
some
additional
events
coming
up
in
June
Washington
DC,
including
a
workshop
on
community
driven
citizen
science
focused
on
health
and
the
environment
and
collaboration
theme,
an
association
to
the
advancement
of
science.
We
will
have
a
big
data
hub
panel
on
Picasso
conference,
so
looking
at
research
collaborations
between
the
US
big
data
research
community
and
the
European
big
data
research
community,
particularly
some
of
the
private
sector.
A
B
So
also,
this
summer
were
running
a
data,
a
program
that
was
part
of
a
series
called
keeping
data
science
fraud
that
people
can
look
at
on
our
YouTube
channel
as
well,
that
works
with
community
colleges,
small
liberal
arts,
schools
and
technical
colleges
to
look
at
their
upskill
and
capacity
for
data
science.
So
some
of
these
things
are
you
talking
about
born
or
those
that
are
interested
in
the
future
of
work.
B
We
want
to
expand
that
group
to
really
get
them
more
and
engaged
so
we're
providing
data
designs,
training,
seven
universities-
this
you
know
this
year
as
well
as
training
in
November.
That
will
coincide
with
getting
faculty
up
skill
to
teach
data
science
in
their
particular
setting.
So
I
think
this
discussion
on
the
future
of
work
is
really
relevant,
especially
to
that
community
as
well
so
I
hope
to
see
you
there
or
hope
to
see
you
as
part
of
that
initiative
as
well
great.
B
So
that
will
happen
in
the
fall.
There's
a
program
they're
running
for
faculty
to
do
fellowships
with
industry,
so
short-term
fellowships
with
industry
and
now
government.
So
it's
Pepe
jeans
know
where
you
can
provide
short-term
pants
or
you
work
with
inside
of
the
company
or
inside
of
the
government
agency
to
really
get
a
sense
of
their
data
structure
and
their
data
needs
to
further
your
research
and
that
relationship.
So
it
had
something
to
look
out
for
so
great.
A
And
those
of
you
who
are
joining
us
new
to
the
hub,
we
have
up
here
on
the
slide,
how
you
think
participate.
You
can
email
us
and
also
get
on
our
list.
We
have
a
monthly
newsletter
with
all
these
kinds
of
opportunities
that
we've
talked
about
today.
The
Twitter
accounts.
If
you
follow
us
on
Twitter
and
if
you
want
to
follow
up
with
us
steroids
of
NSF
grants
program
called
the
spoke.
Big
data
spoke
program
which
should
be
announced
that
needs
sometime
in
the
fall
Oracle.