<s>
In	O
quantum	B-Architecture
computing	I-Architecture
,	O
quantum	B-Device
supremacy	I-Device
,	O
quantum	O
primacy	O
or	O
quantum	B-Device
advantage	I-Device
is	O
the	O
goal	O
of	O
demonstrating	O
that	O
a	O
programmable	O
quantum	O
device	O
can	O
solve	O
a	O
problem	O
that	O
no	O
classical	O
computer	O
can	O
solve	O
in	O
any	O
feasible	O
amount	O
of	O
time	O
(	O
irrespective	O
of	O
the	O
usefulness	O
of	O
the	O
problem	O
)	O
.	O
</s>
<s>
Conceptually	O
,	O
quantum	B-Device
supremacy	I-Device
involves	O
both	O
the	O
engineering	O
task	O
of	O
building	O
a	O
powerful	O
quantum	B-Architecture
computer	I-Architecture
and	O
the	O
computational-complexity-theoretic	O
task	O
of	O
finding	O
a	O
problem	O
that	O
can	O
be	O
solved	O
by	O
that	O
quantum	B-Architecture
computer	I-Architecture
and	O
has	O
a	O
superpolynomial	O
speedup	O
over	O
the	O
best	O
known	O
or	O
possible	O
classical	O
algorithm	B-Algorithm
for	O
that	O
task	O
.	O
</s>
<s>
The	O
term	O
was	O
coined	O
by	O
John	O
Preskill	O
in	O
2012	O
,	O
but	O
the	O
concept	O
of	O
a	O
qualitative	O
quantum	O
computational	O
advantage	O
,	O
specifically	O
for	O
simulating	O
quantum	O
systems	O
,	O
dates	O
back	O
to	O
Yuri	O
Manin	O
's	O
(	O
1980	O
)	O
and	O
Richard	O
Feynman	O
's	O
(	O
1981	O
)	O
proposals	O
of	O
quantum	B-Architecture
computing	I-Architecture
.	O
</s>
<s>
Examples	O
of	O
proposals	O
to	O
demonstrate	O
quantum	B-Device
supremacy	I-Device
include	O
the	O
boson	O
sampling	O
proposal	O
of	O
Aaronson	O
and	O
Arkhipov	O
,	O
D-Wave	O
'	O
s	O
specialized	O
frustrated	O
cluster	O
loop	O
problems	O
,	O
and	O
sampling	O
the	O
output	O
of	O
random	O
quantum	B-Application
circuits	I-Application
.	O
</s>
<s>
A	O
notable	O
property	O
of	O
quantum	B-Device
supremacy	I-Device
is	O
that	O
it	O
can	O
be	O
feasibly	O
achieved	O
by	O
near-term	O
quantum	B-Architecture
computers	I-Architecture
,	O
since	O
it	O
does	O
not	O
require	O
a	O
quantum	B-Architecture
computer	I-Architecture
to	O
perform	O
any	O
useful	O
task	O
or	O
use	O
high-quality	O
quantum	B-Device
error	I-Device
correction	I-Device
,	O
both	O
of	O
which	O
are	O
long-term	O
goals	O
.	O
</s>
<s>
Consequently	O
,	O
researchers	O
view	O
quantum	B-Device
supremacy	I-Device
as	O
primarily	O
a	O
scientific	O
goal	O
,	O
with	O
relatively	O
little	O
immediate	O
bearing	O
on	O
the	O
future	O
commercial	O
viability	O
of	O
quantum	B-Architecture
computing	I-Architecture
.	O
</s>
<s>
Due	O
to	O
unpredictable	O
possible	O
improvements	O
in	O
classical	O
computers	O
and	O
algorithms	O
,	O
quantum	B-Device
supremacy	I-Device
may	O
be	O
temporary	O
or	O
unstable	O
,	O
placing	O
possible	O
achievements	O
under	O
significant	O
scrutiny	O
.	O
</s>
<s>
Turing	O
's	O
paper	O
described	O
what	O
he	O
called	O
a	O
“	O
universal	B-Architecture
computing	I-Architecture
machine	I-Architecture
”	O
,	O
which	O
later	O
became	O
known	O
as	O
a	O
Turing	B-Architecture
machine	I-Architecture
.	O
</s>
<s>
In	O
1980	O
,	O
Paul	O
Benioff	O
utilized	O
Turing	O
's	O
paper	O
to	O
propose	O
the	O
theoretical	O
feasibility	O
of	O
Quantum	B-Architecture
Computing	I-Architecture
.	O
</s>
<s>
His	O
paper	O
,	O
“	O
The	O
Computer	O
as	O
a	O
Physical	O
System	O
:	O
A	O
Microscopic	O
Quantum	O
Mechanical	O
Hamiltonian	O
Model	O
of	O
Computers	O
as	O
Represented	O
by	O
Turing	B-Architecture
Machines	I-Architecture
“	O
,	O
was	O
the	O
first	O
to	O
demonstrate	O
that	O
it	O
is	O
possible	O
to	O
show	O
the	O
reversible	O
nature	O
of	O
quantum	B-Architecture
computing	I-Architecture
as	O
long	O
as	O
the	O
energy	O
dissipated	O
is	O
arbitrarily	O
small	O
.	O
</s>
<s>
During	O
a	O
lecture	O
,	O
he	O
delivered	O
the	O
famous	O
quote	O
,	O
“	O
Nature	O
is	O
n't	O
classical	O
,	O
dammit	O
,	O
and	O
if	O
you	O
want	O
to	O
make	O
a	O
simulation	O
of	O
nature	O
,	O
you	O
'd	O
better	O
make	O
it	O
quantum	O
mechanical	O
,	O
and	O
by	O
golly	O
it	O
's	O
a	O
wonderful	O
problem	O
,	O
because	O
it	O
does	O
n't	O
look	O
so	O
easy.	O
”	O
Soon	O
after	O
this	O
,	O
David	O
Deutsch	O
produced	O
a	O
description	O
for	O
a	O
quantum	O
Turing	B-Architecture
machine	I-Architecture
and	O
designed	O
an	O
algorithm	B-Algorithm
created	O
to	O
run	O
on	O
a	O
quantum	B-Architecture
computer	I-Architecture
.	O
</s>
<s>
In	O
1994	O
,	O
further	O
progress	O
toward	O
quantum	B-Device
supremacy	I-Device
was	O
made	O
when	O
Peter	O
Shor	O
formulated	O
Shor	B-Algorithm
's	I-Algorithm
algorithm	I-Algorithm
,	O
streamlining	O
a	O
method	O
for	O
factoring	O
integers	O
in	O
polynomial	O
time	O
.	O
</s>
<s>
In	O
1996	O
,	O
Lov	O
Grover	O
put	O
into	O
motion	O
an	O
interest	O
in	O
fabricating	O
a	O
quantum	B-Architecture
computer	I-Architecture
after	O
publishing	O
his	O
algorithm	B-Algorithm
,	O
Grover	B-Algorithm
's	I-Algorithm
Algorithm	I-Algorithm
,	O
in	O
his	O
paper	O
,	O
“	O
A	O
fast	O
quantum	O
mechanical	O
algorithm	B-Algorithm
for	O
database	O
search	O
”	O
.	O
</s>
<s>
In	O
1998	O
,	O
Jonathan	O
A	O
.	O
Jones	O
and	O
Michele	O
Mosca	O
published	O
“	O
Implementation	O
of	O
a	O
Quantum	B-Device
Algorithm	I-Device
to	O
Solve	O
Deutsch	B-Algorithm
's	I-Algorithm
Problem	I-Algorithm
on	O
a	O
Nuclear	O
Magnetic	O
Resonance	O
Quantum	B-Architecture
Computer	I-Architecture
”	O
,	O
marking	O
the	O
first	O
demonstration	O
of	O
a	O
quantum	B-Device
algorithm	I-Device
.	O
</s>
<s>
Vast	O
progress	O
toward	O
quantum	B-Device
supremacy	I-Device
was	O
made	O
in	O
the	O
2000s	O
from	O
the	O
first	O
5-qubit	O
nuclear	O
magnetic	O
resonance	O
computer	O
(	O
2000	O
)	O
,	O
the	O
demonstration	O
of	O
Shor	O
's	O
theorem	O
(	O
2001	O
)	O
,	O
and	O
the	O
implementation	O
of	O
Deutsch	B-Algorithm
's	I-Algorithm
algorithm	I-Algorithm
in	O
a	O
clustered	O
quantum	B-Architecture
computer	I-Architecture
(	O
2007	O
)	O
.	O
</s>
<s>
In	O
2011	O
,	O
D-Wave	O
Systems	O
of	O
Burnaby	O
in	O
British	O
Columbia	O
became	O
the	O
first	O
company	O
to	O
sell	O
a	O
quantum	B-Architecture
computer	I-Architecture
commercially	O
.	O
</s>
<s>
In	O
2012	O
,	O
physicist	O
Nanyang	O
Xu	O
landed	O
a	O
milestone	O
accomplishment	O
by	O
using	O
an	O
improved	O
adiabatic	O
factoring	O
algorithm	B-Algorithm
to	O
factor	O
143	O
.	O
</s>
<s>
Not	O
long	O
after	O
this	O
accomplishment	O
,	O
Google	B-Application
purchased	O
its	O
first	O
quantum	B-Architecture
computer	I-Architecture
.	O
</s>
<s>
Google	B-Application
had	O
announced	O
plans	O
to	O
demonstrate	O
quantum	B-Device
supremacy	I-Device
before	O
the	O
end	O
of	O
2017	O
with	O
an	O
array	O
of	O
49	O
superconducting	O
qubits	O
.	O
</s>
<s>
In	O
October	O
2017	O
,	O
IBM	O
demonstrated	O
the	O
simulation	O
of	O
56	O
qubits	O
on	O
a	O
classical	O
supercomputer	B-Architecture
,	O
thereby	O
increasing	O
the	O
computational	O
power	O
needed	O
to	O
establish	O
quantum	B-Device
supremacy	I-Device
.	O
</s>
<s>
In	O
November	O
2018	O
,	O
Google	B-Application
announced	O
a	O
partnership	O
with	O
NASA	O
that	O
would	O
"	O
analyze	O
results	O
from	O
quantum	B-Application
circuits	I-Application
run	O
on	O
Google	B-Application
quantum	B-Architecture
processors	I-Architecture
,	O
and	O
...	O
provide	O
comparisons	O
with	O
classical	O
simulation	O
to	O
both	O
support	O
Google	B-Application
in	O
validating	O
its	O
hardware	O
and	O
establish	O
a	O
baseline	O
for	O
quantum	O
supremacy.	O
"	O
</s>
<s>
Theoretical	O
work	O
published	O
in	O
2018	O
suggests	O
that	O
quantum	B-Device
supremacy	I-Device
should	O
be	O
possible	O
with	O
a	O
"	O
two-dimensional	O
lattice	O
of	O
7×7	O
qubits	O
and	O
around	O
40	O
clock	O
cycles	O
"	O
if	O
error	O
rates	O
can	O
be	O
pushed	O
low	O
enough	O
.	O
</s>
<s>
On	O
June	O
18	O
,	O
2019	O
,	O
Quanta	O
Magazine	O
suggested	O
that	O
quantum	B-Device
supremacy	I-Device
could	O
happen	O
in	O
2019	O
,	O
according	O
to	O
Neven	O
's	O
law	O
.	O
</s>
<s>
On	O
September	O
20	O
,	O
2019	O
,	O
the	O
Financial	O
Times	O
reported	O
that	O
"	O
Google	B-Application
claims	O
to	O
have	O
reached	O
quantum	B-Device
supremacy	I-Device
with	O
an	O
array	O
of	O
54	O
qubits	O
out	O
of	O
which	O
53	O
were	O
functional	O
,	O
which	O
were	O
used	O
to	O
perform	O
a	O
series	O
of	O
operations	O
in	O
200	O
seconds	O
that	O
would	O
take	O
a	O
supercomputer	B-Architecture
about	O
10,000	O
years	O
to	O
complete	O
"	O
.	O
</s>
<s>
On	O
October	O
23	O
,	O
Google	B-Application
officially	O
confirmed	O
the	O
claims	O
.	O
</s>
<s>
IBM	O
responded	O
by	O
suggesting	O
some	O
of	O
the	O
claims	O
are	O
excessive	O
and	O
suggested	O
that	O
it	O
could	O
take	O
2.5	O
days	O
instead	O
of	O
10,000	O
years	O
,	O
listing	O
techniques	O
that	O
a	O
classical	O
supercomputer	B-Architecture
may	O
use	O
to	O
maximize	O
computing	O
speed	O
.	O
</s>
<s>
IBM	O
's	O
response	O
is	O
relevant	O
as	O
the	O
most	O
powerful	O
supercomputer	B-Architecture
at	O
the	O
time	O
,	O
Summit	B-Device
,	O
was	O
made	O
by	O
IBM	O
.	O
</s>
<s>
Researchers	O
have	O
since	O
developed	O
better	O
algorithms	O
for	O
the	O
sampling	O
problem	O
used	O
to	O
claim	O
quantum	B-Device
supremacy	I-Device
,	O
giving	O
substantial	O
reductions	O
to	O
the	O
gap	O
between	O
Sycamore	B-Device
and	O
classical	O
supercomputers	B-Architecture
and	O
even	O
beating	O
it	O
.	O
</s>
<s>
In	O
December	O
2020	O
,	O
a	O
group	O
based	O
in	O
the	O
University	O
of	O
Science	O
and	O
Technology	O
of	O
China	O
(	O
USTC	O
)	O
led	O
by	O
Jian-Wei	O
Pan	O
reached	O
quantum	B-Device
supremacy	I-Device
by	O
implementing	O
gaussian	O
boson	O
sampling	O
on	O
76	O
photons	B-Application
with	O
their	O
photonic	O
quantum	B-Architecture
computer	I-Architecture
Jiuzhang	O
.	O
</s>
<s>
The	O
paper	O
states	O
that	O
to	O
generate	O
the	O
number	O
of	O
samples	O
the	O
quantum	B-Architecture
computer	I-Architecture
generates	O
in	O
20	O
seconds	O
,	O
a	O
classical	O
supercomputer	B-Architecture
would	O
require	O
600	O
million	O
years	O
of	O
computation	O
.	O
</s>
<s>
In	O
October	O
2021	O
,	O
teams	O
from	O
USTC	O
again	O
reported	O
quantum	O
primacy	O
by	O
building	O
two	O
supercomputers	B-Architecture
called	O
Jiuzhang	O
2.0	O
and	O
Zuchongzhi	O
.	O
</s>
<s>
The	O
light-based	O
Jiuzhang	O
2.0	O
implemented	O
gaussian	O
boson	O
sampling	O
to	O
detect	O
113	O
photons	B-Application
from	O
a	O
144-mode	O
optical	O
interferometer	O
and	O
a	O
sampling	O
rate	O
speed	O
up	O
of	O
–	O
a	O
difference	O
of	O
37	O
photons	B-Application
and	O
10	O
orders	O
of	O
magnitude	O
over	O
the	O
previous	O
Jiuzhang	O
.	O
</s>
<s>
Zuchongzhi	O
is	O
a	O
programmable	O
superconducting	O
quantum	B-Architecture
computer	I-Architecture
that	O
needs	O
to	O
be	O
kept	O
at	O
extremely	O
low	O
temperatures	O
to	O
work	O
efficiently	O
and	O
uses	O
random	O
circuit	O
sampling	O
to	O
obtain	O
56	O
qubits	O
from	O
a	O
tunable	O
coupling	O
architecture	O
of	O
66	O
transmons	O
—	O
an	O
improvement	O
over	O
Google	B-Application
's	I-Application
Sycamore	B-Device
2019	O
achievement	O
by	O
3	O
qubits	O
,	O
meaning	O
a	O
greater	O
computational	O
cost	O
of	O
classical	O
simulation	O
of	O
2	O
to	O
3	O
orders	O
of	O
magnitude	O
.	O
</s>
<s>
A	O
third	O
study	O
reported	O
that	O
Zuchongzhi	O
2.1	O
completed	O
a	O
sampling	O
task	O
that	O
"	O
is	O
about	O
6	O
orders	O
of	O
magnitude	O
more	O
difficult	O
than	O
that	O
of	O
Sycamore	B-Device
"	O
"	O
in	O
the	O
classic	O
simulation	O
"	O
.	O
</s>
<s>
In	O
June	O
2022	O
Xanadu	O
has	O
reported	O
on	O
a	O
boson	O
sampling	O
experiment	O
summing	O
up	O
to	O
those	O
of	O
Google	B-Application
and	O
USTC	O
,	O
their	O
setup	O
used	O
loops	O
of	O
optical	O
fiber	O
and	O
multiplexing	B-Architecture
to	O
replace	O
the	O
network	O
of	O
beam	O
splitters	O
by	O
a	O
single	O
one	O
which	O
made	O
it	O
also	O
more	O
easily	O
reconfigurable	O
.	O
</s>
<s>
They	O
detected	O
a	O
mean	O
of	O
125	O
up	O
to	O
219	O
photon	B-Application
from	O
216	O
squeezed	O
modes	O
(	O
squeezed	O
light	O
follows	O
a	O
photon	B-Application
number	O
distribution	O
so	O
they	O
can	O
contain	O
more	O
than	O
one	O
photon	B-Application
per	O
mode	O
)	O
and	O
claim	O
to	O
have	O
obtained	O
a	O
speedup	O
50	O
million	O
times	O
bigger	O
than	O
previous	O
experiments	O
.	O
</s>
<s>
As	O
an	O
extension	O
of	O
classical	O
computational	O
complexity	O
theory	O
,	O
quantum	O
complexity	O
theory	O
considers	O
what	O
a	O
theoretical	O
universal	O
quantum	B-Architecture
computer	I-Architecture
could	O
accomplish	O
without	O
accounting	O
for	O
the	O
difficulty	O
of	O
building	O
a	O
physical	O
quantum	B-Architecture
computer	I-Architecture
or	O
dealing	O
with	O
decoherence	O
and	O
noise	O
.	O
</s>
<s>
Since	O
quantum	O
information	O
is	O
a	O
generalization	O
of	O
classical	O
information	O
,	O
quantum	B-Architecture
computers	I-Architecture
can	O
simulate	O
any	O
classical	O
algorithm	B-Algorithm
.	O
</s>
<s>
The	O
most	O
useful	O
quantum	O
complexity	O
class	O
is	O
BQP	O
(	O
bounded-error	O
quantum	O
polynomial	O
time	O
)	O
,	O
the	O
class	O
of	O
decision	O
problems	O
that	O
can	O
be	O
solved	O
in	O
polynomial	O
time	O
by	O
a	O
universal	O
quantum	B-Architecture
computer	I-Architecture
.	O
</s>
<s>
The	O
difficulty	O
of	O
proving	O
what	O
cannot	O
be	O
done	O
with	O
classical	O
computing	O
is	O
a	O
common	O
problem	O
in	O
definitively	O
demonstrating	O
quantum	B-Device
supremacy	I-Device
.	O
</s>
<s>
If	O
there	O
is	O
a	O
classical	O
algorithm	B-Algorithm
that	O
can	O
efficiently	O
sample	O
from	O
the	O
output	O
of	O
an	O
arbitrary	O
quantum	B-Application
circuit	I-Application
,	O
the	O
polynomial	O
hierarchy	O
would	O
collapse	O
to	O
the	O
third	O
level	O
,	O
which	O
is	O
generally	O
considered	O
to	O
be	O
very	O
unlikely	O
.	O
</s>
<s>
The	O
arguments	O
used	O
to	O
reach	O
this	O
conclusion	O
have	O
been	O
extended	O
to	O
IQP	O
Sampling	O
,	O
where	O
only	O
the	O
conjecture	O
that	O
the	O
average	O
-	O
and	O
worst-case	O
complexities	O
of	O
the	O
problem	O
are	O
the	O
same	O
is	O
needed	O
,	O
as	O
well	O
as	O
to	O
Random	O
Circuit	O
Sampling	O
,	O
which	O
is	O
the	O
task	O
replicated	O
by	O
the	B-Application
Google	I-Application
and	O
USTC	O
research	O
groups	O
.	O
</s>
<s>
Such	O
proposals	O
include	O
(	O
1	O
)	O
a	O
well-defined	O
computational	O
problem	O
,	O
(	O
2	O
)	O
a	O
quantum	B-Device
algorithm	I-Device
to	O
solve	O
this	O
problem	O
,	O
(	O
3	O
)	O
a	O
comparison	O
best-case	O
classical	O
algorithm	B-Algorithm
to	O
solve	O
the	O
problem	O
,	O
and	O
(	O
4	O
)	O
a	O
complexity-theoretic	O
argument	O
that	O
,	O
under	O
a	O
reasonable	O
assumption	O
,	O
no	O
classical	O
algorithm	B-Algorithm
can	O
perform	O
significantly	O
better	O
than	O
current	O
algorithms	O
(	O
so	O
the	O
quantum	B-Device
algorithm	I-Device
still	O
provides	O
a	O
superpolynomial	O
speedup	O
)	O
.	O
</s>
<s>
This	O
algorithm	B-Algorithm
finds	O
the	O
prime	O
factorization	O
of	O
an	O
n-bit	O
integer	O
in	O
time	O
whereas	O
the	O
best	O
known	O
classical	O
algorithm	B-Algorithm
requires	O
time	O
and	O
the	O
best	O
upper	O
bound	O
for	O
the	O
complexity	O
of	O
this	O
problem	O
is	O
.	O
</s>
<s>
It	O
can	O
also	O
provide	O
a	O
speedup	O
for	O
any	O
problem	O
that	O
reduces	O
to	O
integer	O
factoring	O
,	O
including	O
the	O
membership	O
problem	O
for	O
matrix	B-Algorithm
groups	I-Algorithm
over	O
fields	O
of	O
odd	O
order	O
.	O
</s>
<s>
This	O
algorithm	B-Algorithm
is	O
important	O
both	O
practically	O
and	O
historically	O
for	O
quantum	B-Architecture
computing	I-Architecture
.	O
</s>
<s>
It	O
was	O
the	O
first	O
polynomial-time	O
quantum	B-Device
algorithm	I-Device
proposed	O
for	O
a	O
real-world	O
problem	O
that	O
is	O
believed	O
to	O
be	O
hard	O
for	O
classical	O
computers	O
.	O
</s>
<s>
Namely	O
,	O
it	O
gives	O
a	O
superpolynomial	O
speedup	O
under	O
the	O
reasonable	O
assumption	O
that	O
RSA	B-Architecture
,	O
a	O
well-established	O
cryptosystem	O
,	O
is	O
secure	O
.	O
</s>
<s>
However	O
,	O
implementing	O
Shor	B-Algorithm
's	I-Algorithm
algorithm	I-Algorithm
for	O
large	O
numbers	O
is	O
infeasible	O
with	O
current	O
technology	O
,	O
so	O
it	O
is	O
not	O
being	O
pursued	O
as	O
a	O
strategy	O
for	O
demonstrating	O
supremacy	O
.	O
</s>
<s>
This	O
computing	O
paradigm	O
based	O
upon	O
sending	O
identical	O
photons	B-Application
through	O
a	O
linear-optical	O
network	O
can	O
solve	O
certain	O
sampling	O
and	O
search	O
problems	O
that	O
,	O
assuming	O
a	O
few	O
complexity-theoretical	O
conjectures	O
(	O
that	O
calculating	O
the	O
permanent	O
of	O
Gaussian	O
matrices	O
is	O
#P	O
-Hard	O
and	O
that	O
the	O
polynomial	O
hierarchy	O
does	O
not	O
collapse	O
)	O
are	O
intractable	O
for	O
classical	O
computers	O
.	O
</s>
<s>
The	O
largest	O
experimental	O
implementation	O
of	O
boson	O
sampling	O
to	O
date	O
had	O
6	O
modes	O
so	O
could	O
handle	O
up	O
to	O
6	O
photons	B-Application
at	O
a	O
time	O
.	O
</s>
<s>
The	O
best	O
proposed	O
classical	O
algorithm	B-Algorithm
for	O
simulating	O
boson	O
sampling	O
runs	O
in	O
time	O
for	O
a	O
system	O
with	O
n	O
photons	B-Application
and	O
m	O
output	O
modes	O
.	O
</s>
<s>
is	O
an	O
open-source	O
implementation	O
in	O
R	B-Language
.	O
The	O
algorithm	B-Algorithm
leads	O
to	O
an	O
estimate	O
of	O
50	O
photons	B-Application
required	O
to	O
demonstrate	O
quantum	B-Device
supremacy	I-Device
with	O
boson	O
sampling	O
.	O
</s>
<s>
The	O
best	O
known	O
algorithm	B-Algorithm
for	O
simulating	O
an	O
arbitrary	O
random	O
quantum	B-Application
circuit	I-Application
requires	O
an	O
amount	O
of	O
time	O
that	O
scales	O
exponentially	O
with	O
the	O
number	O
of	O
qubits	O
,	O
leading	O
one	O
group	O
to	O
estimate	O
that	O
around	O
50	O
qubits	O
could	O
be	O
enough	O
to	O
demonstrate	O
quantum	B-Device
supremacy	I-Device
.	O
</s>
<s>
Bouland	O
,	O
Fefferman	O
,	O
Nirkhe	O
and	O
Vazirani	O
gave	O
,	O
in	O
2018	O
,	O
theoretical	O
evidence	O
that	O
efficiently	O
simulating	O
a	O
random	O
quantum	B-Application
circuit	I-Application
would	O
require	O
a	O
collapse	O
of	O
the	O
computational	O
polynomial	O
hierarchy	O
.	O
</s>
<s>
Google	B-Application
had	O
announced	O
its	O
intention	O
to	O
demonstrate	O
quantum	B-Device
supremacy	I-Device
by	O
the	O
end	O
of	O
2017	O
by	O
constructing	O
and	O
running	O
a	O
49-qubit	O
chip	O
that	O
would	O
be	O
able	O
to	O
sample	O
distributions	O
inaccessible	O
to	O
any	O
current	O
classical	O
computers	O
in	O
a	O
reasonable	O
amount	O
of	O
time	O
.	O
</s>
<s>
The	O
largest	O
universal	O
quantum	B-Application
circuit	I-Application
simulator	O
running	O
on	O
classical	O
supercomputers	B-Architecture
at	O
the	O
time	O
was	O
able	O
to	O
simulate	O
48	O
qubits	O
.	O
</s>
<s>
But	O
for	O
particular	O
kinds	O
of	O
circuits	O
,	O
larger	O
quantum	B-Application
circuit	I-Application
simulations	O
with	O
56	O
qubits	O
are	O
possible	O
.	O
</s>
<s>
This	O
may	O
require	O
increasing	O
the	O
number	O
of	O
qubits	O
to	O
demonstrate	O
quantum	B-Device
supremacy	I-Device
.	O
</s>
<s>
On	O
October	O
23	O
,	O
2019	O
,	O
Google	B-Application
published	O
the	O
results	O
of	O
this	O
quantum	B-Device
supremacy	I-Device
experiment	O
in	O
the	O
Nature	O
article	O
,	O
“	O
Quantum	B-Device
Supremacy	I-Device
Using	O
a	O
Programmable	O
Superconducting	O
Processor	O
”	O
in	O
which	O
they	O
developed	O
a	O
new	O
53-qubit	O
processor	O
,	O
named	O
“	O
Sycamore	B-Device
”	O
,	O
that	O
is	O
capable	O
of	O
fast	O
,	O
high-fidelity	O
quantum	O
logic	O
gates	O
,	O
in	O
order	O
to	O
perform	O
the	O
benchmark	O
testing	O
.	O
</s>
<s>
Google	B-Application
claims	O
that	O
their	O
machine	O
performed	O
the	O
target	O
computation	O
in	O
200	O
seconds	O
,	O
and	O
estimated	O
that	O
their	O
classical	O
algorithm	B-Algorithm
would	O
take	O
10,000	O
years	O
in	O
the	O
world	O
's	O
fastest	O
supercomputer	B-Architecture
to	O
solve	O
the	O
same	O
problem	O
.	O
</s>
<s>
IBM	O
disputed	O
this	O
claim	O
,	O
saying	O
that	O
an	O
improved	O
classical	O
algorithm	B-Algorithm
should	O
be	O
able	O
to	O
solve	O
that	O
problem	O
in	O
two	O
and	O
a	O
half	O
days	O
on	O
that	O
same	O
supercomputer	B-Architecture
.	O
</s>
<s>
Quantum	B-Architecture
computers	I-Architecture
are	O
much	O
more	O
susceptible	O
to	O
errors	O
than	O
classical	O
computers	O
due	O
to	O
decoherence	O
and	O
noise	O
.	O
</s>
<s>
The	O
threshold	O
theorem	O
states	O
that	O
a	O
noisy	O
quantum	B-Architecture
computer	I-Architecture
can	O
use	O
quantum	B-Device
error-correcting	I-Device
codes	I-Device
to	O
simulate	O
a	O
noiseless	O
quantum	B-Architecture
computer	I-Architecture
,	O
assuming	O
the	O
error	O
introduced	O
in	O
each	O
computer	O
cycle	O
is	O
less	O
than	O
some	O
number	O
.	O
</s>
<s>
However	O
,	O
it	O
is	O
not	O
yet	O
definitively	O
known	O
how	O
the	O
resources	O
needed	O
for	O
error	B-Device
correction	I-Device
will	O
scale	O
with	O
the	O
number	O
of	O
qubits	O
.	O
</s>
<s>
Skeptics	O
point	O
to	O
the	O
unknown	O
behavior	O
of	O
noise	O
in	O
scaled-up	O
quantum	O
systems	O
as	O
a	O
potential	O
roadblock	O
for	O
successfully	O
implementing	O
quantum	B-Architecture
computing	I-Architecture
and	O
demonstrating	O
quantum	B-Device
supremacy	I-Device
.	O
</s>
<s>
Some	O
researchers	O
have	O
suggested	O
that	O
the	O
term	O
"	O
quantum	B-Device
supremacy	I-Device
"	O
should	O
not	O
be	O
used	O
,	O
arguing	O
that	O
the	O
word	O
"	O
supremacy	O
"	O
evokes	O
distasteful	O
comparisons	O
to	O
the	O
racist	O
belief	O
of	O
white	O
supremacy	O
.	O
</s>
<s>
A	O
controversial	O
commentary	O
article	O
in	O
the	O
journal	O
Nature	O
signed	O
by	O
thirteen	O
researchers	O
asserts	O
that	O
the	O
alternative	O
phrase	O
"	O
quantum	B-Device
advantage	I-Device
"	O
should	O
be	O
used	O
instead	O
.	O
</s>
<s>
John	O
Preskill	O
,	O
the	O
professor	O
of	O
theoretical	O
physics	O
at	O
the	O
California	O
Institute	O
of	O
Technology	O
who	O
coined	O
the	O
term	O
,	O
has	O
since	O
clarified	O
that	O
the	O
term	O
was	O
proposed	O
to	O
explicitly	O
describe	O
the	O
moment	O
that	O
a	O
quantum	B-Architecture
computer	I-Architecture
gains	O
the	O
ability	O
to	O
perform	O
a	O
task	O
that	O
a	O
classical	O
computer	O
never	O
could	O
.	O
</s>
<s>
He	O
further	O
explained	O
that	O
he	O
specifically	O
rejected	O
the	O
term	O
"	O
quantum	B-Device
advantage	I-Device
"	O
as	O
it	O
did	O
not	O
fully	O
encapsulate	O
the	O
meaning	O
of	O
his	O
new	O
term	O
:	O
the	O
word	O
"	O
advantage	O
"	O
would	O
imply	O
that	O
a	O
computer	O
with	O
quantum	B-Device
supremacy	I-Device
would	O
have	O
a	O
slight	O
edge	O
over	O
a	O
classical	O
computer	O
while	O
the	O
word	O
"	O
supremacy	O
"	O
better	O
conveys	O
complete	O
ascendancy	O
over	O
any	O
classical	O
computer	O
.	O
</s>
<s>
Nature	O
's	O
Philip	O
Ball	O
wrote	O
in	O
December	O
2020	O
that	O
the	O
term	O
"	O
quantum	B-Device
advantage	I-Device
"	O
has	O
"	O
largely	O
replaced	O
"	O
the	O
term	O
"	O
quantum	B-Device
supremacy	I-Device
"	O
.	O
</s>
