Bloomberg Businessweek - USA (2020-11-02)

39

ILLUSTRATION BY 731

◼SOLUTIONS BloombergBusinessweek November 2, 2020

A decadeago,twoscientistsatHoneywellInternationalInc.

pitchedtheirbossesa mostimprobableidea:Thecentury-

oldindustrialcompanyshouldbuilda quantumcomputer.

Theideawasmetwithsomeskepticism.Foronething,

InternationalBusinessMachinesCorp.andGooglewere

alreadymovingaheadonquantumcomputing,whichprom-

isestobemuchfaster,forcertaintasks,thaneventhe

speediesttransistor-basedchips.AndHoneywellwasbet-

terknownformakingcockpitcontrols,securitysystems,

andevenworkboots.

Yetmanagementsignedoff,andscientistssetupa lab-

oratoryin a one-storybrickbuildingoutsideBoulder,Colo.

Now,Honeywellis pursuingitsownpathtoproducinga

practicalquantumcomputerthatcanbeusedforevery-

daybusiness.OnOct. 29 it unveiledwhatit dubbedthe

H1,itssecond-generationquantumcomputer.JPMorgan

Chase& Co.,BPPlc,andothercompaniesarepayingfor

subscriptionstowritealgorithmsonthemachineandtest

thoseresultswithclassicalsupercomputers.

Thesedealsareverymuchexperiments—Honeywell’s

clientsaren’tusingitsquantummachinetomakebusiness

decisions,andanycommercialquantumcomputeris years

away.AndHoneywell’sachievementsarelessimpressive

thanGoogle’sboastlastyearthatit hadachieved“quantum

supremacy”afteritsmachineperformeda calculationthat

chokeseventhelargestsupercomputer.IBMfirstmadea

quantumcomputeravailabletothepublicin 2016andhas

rolledoutregularupgrades.

Honeywellmakesa claimthat’simpressingindustry

observers:Itsmachine,thoughslower,appearstoproduce

moreaccurateresults.In thisearlystageallquantumcom-

putersmakeerrors,ornoise,asscientistscallit. Honeywell

hasbeenabletominimizethem.“Qualityis reallytheholy

grail,”saysDeniseRuffner,whopreviouslyworkedatIBM’s

quantumunitandjuststartedasa vicepresidentatIonQ,

a startupquantumcomputermaker.“Whatanyuserreally

wantsis a quantumcomputerwiththeleastamountof

errorornoise.Sofar,thetechnologythatHoneywellis

usinghasbeenthebestatdoingthat.”

Quantumcomputingis basedonthemind-bendingcon-

ceptthatatomicparticlescanbein multiplestatesatonce.

Conventionalcomputersdependonelectriccurrentsbeing

switchedonoroff—statesthatarerepresentedbya “1”

anda “0.”Quantumcomputersuseunitsknownasquan-

tumbits,orqubits,whichcanrepresentbotha “0”anda

“1” at the same time. Generally, quantum computers look

at all the different combinations of data simultaneously,

while classical computers work sequentially, making them

muchslowerwhendealingwithlargesetsofvariables.

Buterrorsareprevalent.Honeywelldecidedearlyonto

useexistingatomsasthebasisofitscomputer,a technol-

ogyknownastrappedion,tominimizeerrors.Google,IBM,

andothersareassemblingtinydevicesthatsimulateparti-

clesusingelectriccurrentsatextremelycoldtemperatures,

a methodcalledsuperconducting.Thistechnologymakes

fora fastercomputerthatcanbescaledupin powerusing

theindustry’sexistingsupplychain.Butit’salsomoreerror-

prone.“Anydesignelementis lookingata trade-off,”says

IlyasKhan,chiefexecutiveofficerofCambridgeQuantum

Computing,a softwarestartupthatcountsHoneywelland

IBMasinvestors.“Ifyouwanteda speedster,you’dbe

lookingatyoursuperconductingquantummachine.Andif

you’remorelookingataccuracyandyoudidn’tcareabout

speed,you’dbelookingattheiontrap.”

HoneywellCEODariusAdamczykpredictsthatwithin

10 yearsitsquantumcomputingbusinesswillhavesales

of$1billion(thecompanyhadabout$37billioninrev-

enuelastyear).“Overtime,themarketisgoingtobe

worthbillions,”hesays.

Oneskepticis BobSutor,vicepresidentforquantum

solutionsatIBM,whosaysHoneywell’strapped-iontech-

nologywillhavetroublescalingup.“Webelievewecanget

throughthescalingproblem,”hesays.“Theiontrappeople

willfacethat.Theywillhitit, andtheywillhitit prettyhard

in thenot-too-far-offfuture.”

JPMorganChaseisrunningsoftwareprogramson

Honeywell’scomputerwithsmallsetsofdatathatcanbe

checkedwitha standardcomputer,helpingboostconfi-

denceontheaccuracy,saysMarcoPistoia,chiefofthe

bank’sFutureLabforAppliedResearchandEngineering.

Hesayshe’seagertorunmorecomplexalgorithmsasthe

computersbecomemorepowerfulsoJPMorgancanbe

readywhenthemachinesroutinelysolveproblemsthata

standardcomputercan’t.

Both approaches are far from finished. “Super-

conducting qubits have come a long way and today serve

as a proof point of the potential that quantum computing

promises,” says James Sanders, an analyst at S&P Global’s

451 Research. “Proponents of trapped ions tout increased

accuracy in comparison, but these are presently slower

to operate. Both need further development to eclipse the

capabilities of classical computers.” �Thomas Black

Tiny Steps Toward a

Quantum Leap

Honeywell is betting on

a slower but less error-

prone technology

THE BOTTOM LINE Honeywell is carving out a pathway in quantum

computing with slower, more accurate technology compared with industry

front-runners IBM and Google.