👂 🎴 🕸️
<
p
class
='
fragment
'>
We
start
with
a
1966
BBC
programme
''
made
twelve
years
after
his
death
.
p
><
p
class
='
fragment
'>
While
you
watch
''
keep
one
question
in
mind
:
<
strong
>
what
is
this
“
machine
”
made
of
?
strong
>
p
>
<
p
class
='
fragment
'>
Born
in
London
''
23
June
1912
.
Mathematician
:
Cambridge
''
then
Princeton
.
p
><
p
class
='
fragment
'>
1936
''
aged
23
:
<
em
>
On
Computable
Numbers
em
>
—
the
paper
that
defines
the
computer
before
any
computer
exists
.
p
><
p
class
='
fragment
'>
1939
–
45
:
Bletchley
Park
''
breaking
German
naval
Enigma
.
p
><
p
class
='
fragment
'>
1945
–
50
:
designs
an
electronic
computer
;
1950
:
<
em
>
Computing
Machinery
and
Intelligence
em
>.
p
><
p
class
='
fragment
'>
1952
:
convicted
for
homosexual
acts
''
chemically
castrated
.
7
June
1954
:
dead
of
cyanide
poisoning
''
aged
41
.
Royal
pardon
:
2013
.
p
><
p
class
='
fragment
'><
strong
>
The
man
who
asked
whether
machines
can
think
was
not
allowed
to
be
what
he
was
.
strong
>
p
>
<
img
loading
=
„
lazy
“
style
=
„
width
:
auto
;
height
:
61vh
;
float
:
left
;
margin
-
right
:
10px
;
“
src
=
„
/
images
/
prolegomena
/
turing
/
imitation1
.
svg
“
/><
img
loading
=
„
lazy
“
style
=
„
width
:
auto
;
height
:
61vh
;
float
:
right
;
margin
-
right
:
10px
;
“
src
=
„
/
images
/
prolegomena
/
turing
/
imitation2
.
svg
“
/>
<
p
class
='
fragment
'>
Turing
refuses
to
define
“
think
”.
He
calls
the
question
<
em
>“
too
meaningless
to
deserve
discussion
”
em
>
and
replaces
it
with
a
game
.
p
><
p
class
='
fragment
'>
Three
players
:
<
strong
>
A
strong
>
(
a
man
)''
<
strong
>
B
strong
>
(
a
woman
)''
<
strong
>
C
strong
>
(
the
interrogator
''
in
another
room
).
C
types
questions
and
must
tell
who
is
the
woman
.
A
tries
to
deceive
.
B
tries
to
help
.
p
><
p
class
='
fragment
'>
Now
let
a
<
strong
>
machine
strong
>
play
A
.
Does
C
get
it
wrong
as
often
as
before
?
p
><
p
class
='
fragment
'>
His
prediction
:
around
the
year
2000
''
an
average
interrogator
would
have
no
more
than
a
70
%
chance
of
a
right
identification
after
five
minutes
.
p
><
p
class
='
fragment
'><
strong
>
The
game
is
called
<
em
>
imitation
em
>
—
and
it
is
about
gender
''
deception
and
performance
.
strong
>
p
>
<
img
loading
=
„
lazy
“
style
=
„
width
:
100
%;
max
-
height
:
70vh
;
object
-
fit
:
contain
;
“
src
=
„
/
images
/
prolegomena
/
turing
/
turingtest
.
png
“
/><
p
style
=
„
font
-
size
:
32px
;
“
><
i
>
„
A
TT
is
a
way
how
to
address
the
question
“
Can
machines
think
?
„
a
>
in
a
scientifically
plausible
ye
deeply
empathic
way
.
More
concretely
''
Turing
(
1950
)
proposes
that
the
performance
of
an
AA
under
question
shall
be
evaluated
by
a
human
judge
J
whose
objective
is
to
determine
which
among
two
entities
-
with
whom
J
is
in
real
-
time
interaction
-
is
of
human
and
which
is
of
artificial
nature
.
“
i
>
(
Hromada
''
2012
)
p
>
<
p
class
='
fragment
'>
The
famous
question
''
in
video
form
.
p
><
p
class
='
fragment
'>
While
watching
''
note
the
one
objection
you
would
raise
.
p
>
<
p
class
='
fragment
'>
Section
6
of
the
paper
lists
nine
objections
''
among
them
theological
''
mathematical
''
from
consciousness
—
and
Lady
Lovelace
'
s
.
p
><
p
class
='
fragment
'>
Lovelace
(
1843
):
the
Analytical
Engine
<
em
>“
has
no
pretensions
to
originate
anything
.
It
can
do
whatever
we
know
how
to
order
it
to
perform
.”
em
>
p
><
p
class
='
fragment
'>
Turing
:
machines
surprise
me
all
the
time
.
And
instead
of
programming
an
adult
mind
''
build
a
<
strong
>
child
machine
strong
>
and
educate
it
.
p
><
p
class
='
fragment
'><
strong
>
Keep
that
sentence
.
It
is
the
first
mention
of
a
learning
machine
.
strong
>
p
>
<
p
class
='
fragment
'>
ELIZA
(
Weizenbaum
''
1966
):
a
few
pattern
rules
''
and
some
users
confided
in
it
anyway
.
p
><
p
class
='
fragment
'>
The
Chinese
Room
(
Searle
''
1980
):
symbols
shuffled
correctly
''
nothing
understood
.
p
><
p
class
='
fragment
'>
Recent
controlled
studies
report
that
large
language
models
are
judged
to
be
human
about
as
often
as
''
or
more
often
than
''
real
humans
(
Jones
&
Bergen
).
p
><
p
class
='
fragment
'><
strong
>
The
test
measures
our
credulity
as
much
as
the
machine
'
s
intelligence
.
strong
>
p
>
<
p
class
='
fragment
'>
Alan
is
a
<
strong
>
Matrix
bot
strong
>
(<
code
>@
alan
:
udk
.
ai
code
>)
running
on
our
own
server
.
You
will
chat
with
“
Alan
”
—
but
behind
him
may
be
an
AI
persona
<
em
>
or
another
student
em
>.
Alan
relays
every
message
''
so
you
are
never
told
which
is
which
.
p
><
p
class
='
fragment
'><
strong
>
How
to
reach
him
:
strong
>
in
Element
(
with
your
account
on
udk
.
ai
or
medienhaus
.
udk
-
berlin
.
de
)
start
a
direct
message
with
<
code
>@
alan
:
udk
.
ai
code
>
and
type
<
code
>
start
code
>.
p
><
p
class
='
fragment
'>
Alan
pairs
you
with
a
waiting
fellow
student
or
with
an
AI
persona
.
After
a
few
minutes
he
asks
:
<
em
>“
Am
I
a
human
or
AI
?”
em
>
Then
comes
the
reveal
:
which
persona
(
and
model
)
it
was
—
or
''
if
you
spoke
with
a
human
''
who
.
p
><
p
class
='
fragment
'>
Good
to
know
:
Alan
cannot
read
encrypted
chats
''
so
use
a
normal
unencrypted
DM
(
if
you
invite
him
to
an
encrypted
one
''
he
leaves
and
opens
a
fresh
DM
with
you
).
The
conversations
are
stored
for
the
course
evaluation
—
don
'
t
share
anything
private
.
p
><
p
class
='
fragment
'>
Try
Turing
'
s
own
questions
:
<
em
>“
Please
write
me
a
sonnet
on
the
Forth
Bridge
.”
em
>
<
em
>“
Add
34957
to
70764
.”
em
>
p
><
p
class
='
fragment
'>
Decide
:
human
or
machine
?
Write
down
the
<
strong
>
three
cues
strong
>
that
decided
it
.
p
><
p
class
='
fragment
'><
strong
>
Turing
'
s
machine
answered
105621
.
The
right
answer
is
105721
.
strong
>
p
>
<
p
class
='
fragment
'>
Which
cues
did
you
use
?
Speed
?
Spelling
?
Politeness
?
Never
saying
“
I
don
'
t
know
”?
A
body
''
a
weather
''
a
bad
day
?
p
><
p
class
='
fragment
'>
Humans
make
typos
and
slip
.
Turing
wrote
that
a
machine
should
<
em
>
deliberately
introduce
mistakes
em
>
to
be
convincing
.
p
><
p
class
='
fragment
'>
Which
cues
tell
you
about
the
machine
''
and
which
only
about
<
em
>
your
expectations
of
humans
em
>?
p
><
p
class
='
fragment
'><
strong
>
Back
to
your
criterion
from
the
Question
of
the
Day
:
did
Alan
meet
it
?
strong
>
p
>
<
p
class
='
fragment
'>
Enigma
:
a
German
cipher
machine
.
Press
a
key
''
a
different
letter
lights
up
—
and
the
rotors
turn
''
so
the
next
key
is
scrambled
differently
.
p
><
p
class
='
fragment
'>
Bletchley
Park
:
Turing
designs
the
<
strong
>
Bombe
strong
>''
a
machine
that
hunts
for
the
day
'
s
settings
(
building
on
Polish
work
).
p
><
p
class
='
fragment
'>
A
machine
breaking
a
machine
.
p
><
p
class
='
fragment
'><
strong
>
And
<
em
>
enigma
em
>
is
also
what
we
call
a
riddle
.
Keep
both
meanings
in
mind
.
strong
>
p
>
<
p
class
='
fragment
'>
1928
''
David
Hilbert
:
is
there
a
<
strong
>
mechanical
procedure
strong
>
that
decides
''
for
any
mathematical
statement
''
whether
it
is
provable
?
p
><
p
class
='
fragment
'>
Leibniz
had
dreamed
of
it
:
<
em
>
let
us
calculate
em
>.
p
><
p
class
='
fragment
'>
1931
''
Kurt
Gödel
:
every
consistent
formal
system
has
true
statements
it
cannot
prove
.
p
><
p
class
='
fragment
'>
1936
''
Alonzo
Church
and
Alan
Turing
''
independently
:
<
strong
>
no
such
procedure
exists
strong
>.
p
><
p
class
='
fragment
'><
strong
>
But
to
prove
that
nothing
can
be
done
''
you
first
need
to
say
exactly
what
“
a
procedure
”
is
.
strong
>
p
>
<
p
class
='
fragment
'>
Until
the
1940s
a
“
computer
”
was
a
<
em
>
person
em
>
who
computes
—
often
a
woman
with
pencil
and
paper
at
a
desk
.
p
><
p
class
='
fragment
'>
Turing
asks
:
what
does
such
a
person
actually
do
?
p
><
p
class
='
fragment
'>
Looks
at
a
symbol
.
Writes
a
symbol
.
Changes
their
state
of
mind
.
Moves
on
to
the
next
square
.
p
><
p
class
='
fragment
'>
Strip
away
everything
else
:
p
><
p
class
='
fragment
'><
strong
>
a
tape
''
a
head
''
a
finite
list
of
rules
.
strong
>
p
>
A
„
theoretical
“
machine
consisting
of
<
img
loading
=
„
lazy
“
style
=
„
width
:
auto
;
height
:
70vh
;
float
:
left
;
margin
-
right
:
50px
;
“
src
=
„
/
images
/
prolegomena
/
turing
/
machine
.
gif
“
/><
ul
style
=
„
margin
-
left
:
80px
;
“
><
li
>
tape
with
discrete
symbols
<
li
>
reading
/
writing
header
<
li
><
a
href
=
„
/
prolegomena
/
repetitio
/
index
.
html
#
binary
adder
fingers
“
target
=
„
blank
“
>
table
of
rules
a
><
li
>
tape
-
moving
mechanism
li
>
ul
><
div
class
=
„
copyfck
“
style
=
„
left
:
5vw
;
top
:
110vh
;
“
>
Visualization
of
the
Turing
Machine
(
Rosario
van
Tulpe
''
2007
''
CC
BY
-
SA
)
div
>
<
p
class
='
fragment
'>
The
tape
holds
a
binary
number
:
<
code
>
1
0
1
1
code
>.
The
head
stands
on
the
<
strong
>
rightmost
strong
>
digit
.
State
:
<
code
>
CARRY
code
>.
p
><
p
class
='
fragment
'><
code
>
CARRY
code
>
reads
<
code
>
1
code
>
→
write
<
code
>
0
code
>''
move
left
''
stay
in
<
code
>
CARRY
code
>
p
><
p
class
='
fragment
'><
code
>
CARRY
code
>
reads
<
code
>
0
code
>
or
blank
→
write
<
code
>
1
code
>''
<
strong
>
STOP
strong
>
p
><
p
class
='
fragment
'>
Run
it
with
pencil
and
paper
.
What
is
on
the
tape
when
it
stops
?
Now
try
<
code
>
1
1
1
code
>.
p
><
p
class
='
fragment
'>
It
adds
one
:
1011
→
1100
''
111
→
1000
.
Three
rules
''
no
understanding
.
p
><
p
class
='
fragment
'><
strong
>
Anything
you
would
call
an
algorithm
can
be
written
as
such
a
table
(
Church
–
Turing
thesis
).
strong
>
p
>
<
p
class
='
fragment
'>
Counting
from
0
to
31
on
the
fingers
of
one
hand
<
em
>
is
em
>
a
Turing
machine
at
work
.
Here
is
its
anatomy
:
p
><
p
class
='
fragment
'><
strong
>
Tape
strong
>
—
your
five
fingers
:
thumb
=
rightmost
cell
''
little
finger
=
leftmost
.
<
strong
>
Symbols
strong
>
—
<
code
>
1
code
>
finger
up
''
<
code
>
0
code
>
finger
down
.
p
><
p
class
='
fragment
'><
strong
>
Head
strong
>
—
the
finger
you
are
looking
at
;
it
starts
on
the
thumb
.
<
strong
>
States
strong
>
—
<
code
>
CARRY
code
>
(
start
)
and
<
code
>
HALT
code
>.
p
><
p
class
='
fragment
'><
strong
>
Transition
rules
strong
><
br
/><
code
>
CARRY
code
>
reads
<
code
>
1
code
>
→
write
<
code
>
0
code
>
(
fold
it
down
)''
move
left
''
stay
in
<
code
>
CARRY
code
><
br
/><
code
>
CARRY
code
>
reads
<
code
>
0
code
>
→
write
<
code
>
1
code
>
(
raise
it
)''
<
code
>
HALT
code
>
p
><
p
class
='
fragment
'><
strong
>
One
run
=
one
count
.
strong
>
Take
7
→
8
:
tape
<
code
>
0
0
1
1
1
code
>.
Thumb
1
→
0
''
index
1
→
0
''
middle
1
→
0
''
ring
0
→
1
''
halt
:
<
code
>
0
1
0
0
0
code
>.
p
><
p
class
='
fragment
'>
Counting
to
31
means
starting
the
machine
31
times
(
57
steps
in
all
).
Same
two
rules
every
time
;
nothing
in
it
“
knows
”
what
a
number
is
.
Adding
two
numbers
on
two
hands
is
the
same
idea
with
a
bigger
rule
table
.
p
><
p
class
='
fragment
'><
strong
>
Now
count
up
from
31
:
strong
>
the
head
falls
off
the
left
end
.
Turing
'
s
tape
is
<
em
>
infinite
em
>;
your
hand
is
not
.
Borrow
your
other
hand
—
or
paper
—
and
keep
going
.
p
>
„
<
i
>
It
is
possible
to
invent
a
single
machine
which
can
be
used
to
compute
any
computable
sequence
.
If
this
machine
U
is
supplied
with
a
tape
on
the
beginning
of
which
is
written
the
S
.
D
[
“
standard
description
„
of
an
action
table
]
of
some
computing
machine
M
''
then
U
will
compute
the
same
sequence
as
M
i
>.
“
(
Turing
''
1936
)
<
p
class
='
fragment
'>
A
universal
machine
reads
the
rules
of
<
em
>
another
em
>
machine
from
its
tape
and
behaves
like
it
.
p
><
p
class
='
fragment
'>
Program
=
data
.
Your
laptop
is
such
a
machine
;
an
app
is
just
a
long
tape
.
p
><
p
class
='
fragment
'><
strong
>
And
a
program
that
treats
programs
as
data
can
also
<
em
>
change
em
>
them
.
Hold
on
to
this
for
the
learning
machines
.
strong
>
p
>
<
div
class
=
„
slidetitle
“
style
=
„
font
-
size
:
40px
;
“
><
a
href
=
„
/
handouts
/
turing
-
on
-
computable
-
numbers
.
pdf
“
target
=
„
blank
“
>
On
Computable
Numbers
''
with
an
Application
to
the
„
Entscheidungsproblem
“
(
1936
)
a
>
div
><
ul
style
=
„
margin
-
left
:
80px
;
font
-
size
:
36px
;
“
><
li
><
b
>
halting
problem
b
>
is
the
problem
of
determining
''
from
a
description
of
an
arbitrary
computer
program
and
an
input
''
whether
the
program
will
finish
running
(
i
.
e
.''
halt
)
or
continue
to
run
forever
<
li
>
Turing
proved
that
a
general
algorithm
to
solve
the
halting
problem
for
all
possible
program
-
input
pairs
cannot
exist
<
li
>
to
prove
this
''
he
described
a
paradox
involving
a
concept
now
known
as
„
Turing
machine
“
ul
>
<
p
class
='
fragment
'>
Suppose
a
program
<
code
>
H
(
p
''
x
)
code
>
always
answers
correctly
:
<
em
>“
p
halts
on
x
”
em
>
or
<
em
>“
p
loops
forever
”
em
>.
p
><
p
class
='
fragment
'>
Build
<
code
>
D
(
p
)
code
>:
ask
<
code
>
H
(
p
''
p
)
code
>.
If
it
says
<
em
>
halts
em
>''
loop
forever
.
If
it
says
<
em
>
loops
em
>''
stop
.
p
><
p
class
='
fragment
'>
Run
<
code
>
D
code
>
on
itself
.
Halts
if
and
only
if
it
loops
.
p
><
p
class
='
fragment
'><
strong
>
Contradiction
:
<
code
>
H
code
>
cannot
exist
.
Some
questions
about
programs
no
program
can
answer
.
strong
>
p
>
<
p
class
='
fragment
'>
Turn
to
your
neighbour
.
Two
minutes
.
p
><
p
class
='
fragment
'>
What
would
a
machine
have
to
<
em
>
do
em
>
for
you
to
say
“
yes
''
it
thinks
”?
p
><
p
class
='
fragment
'>
Write
the
criterion
down
.
p
><
p
class
='
fragment
'><
strong
>
We
will
test
it
at
the
end
of
the
session
.
strong
>
p
>
<
div
style
=
„
font
-
size
:
60px
;
“
><
i
>“
We
may
say
that
thinking
is
essentially
the
activity
of
operating
with
signs
.”
i
>
div
><
br
/><
div
style
=
„
font
-
size
:
32px
;
“
>—
Ludwig
Wittgenstein
''
<
i
>
The
Blue
Book
i
>
(
1933
–
1934
)''
p
. 
6
of
the
commonly
used
edition
div
>
<
i
><
a
href
=
„
/
handouts
/
turing
-
essentials
.
pdf
“
target
=
„
blank
“
>
I
propose
to
consider
the
question
''
„
Can
machines
think
?
“
This
should
begin
with
defnitions
of
the
meaning
of
the
terms
„
machine
“
and
„
think
“
.
a
>
i
>(
Turing
''
1950
)
<
p
class
='
fragment
'><
strong
>
Turing
machine
strong
>:
what
can
be
computed
—
defined
by
what
a
procedure
<
em
>
does
em
>''
not
what
it
is
made
of
.
p
><
p
class
='
fragment
'><
strong
>
Turing
test
strong
>:
what
we
will
call
intelligent
—
defined
by
how
it
behaves
toward
us
.
p
><
p
class
='
fragment
'>
The
question
moves
from
“
what
is
it
?”
to
“
what
does
it
do
with
us
?”
p
><
p
class
='
fragment
'><
strong
>
Next
:
Claude
''
Marvin
''
John
&
c
.
—
from
information
theory
to
Dartmouth
.
strong
>
p
>
[Impressum, Datenschutz, Login] Other subprojects of udk.ai linkring: baumhaus.digital gardens.digital teacher.solar fibel.digital refused.science