A few weeks ago I woke up at a very early hour in Brooklyn. I hopped in my car and drove down
the Hudson River to the small Westchester County community of Yorktown Heights. The town is
surrounded by fields with old farmsteads lies the Thomas J. Watson Research Center The Eero
Saarinen designed the 1960s Jet Age headquarters of IBM Research.
Inside that building is a maze of pathways and security gates that are guarded by the iris
scanner. IBM’s scientists are at work in the development of their ideas for what IBM research
director Dario Gil has told me is “the next branch of computing” quantum computers.
I attended The Watson Center to preview IBM’s updated roadmap for large-scale quantum
computing that is practical and practical. There was a lot of discussion concerning “qubit count,”
“quantum coherence,” “error mitigation,” “software orchestration” and other subjects. You’ll
require to be an electrical engineer who has an education in computer science as well as a
knowledge of quantum mechanics in order to be able to follow.
I’m not one of those, however I have observed the quantum computing field for the past long
enough to be aware that the work being conducted by IBM researchers as well as their
counterparts at companies such as Google and Microsoft as well as many startups from around
the world will be driving the next big quantum leap in computer technology. That, considering the fact that computing is one of the “horizontal technologies that touches everything,” as Gil
explained to me, could have significant impacts on the development of everything that including
security as well as Artificial Intelligence to create better batteries.
If, of course, they actually manage to get these things to work.
Quantum realms: entering the quantum realm
The most effective way to comprehend quantum computers — other than taking a few years off
to attend graduate studies or a post-grad program at MIT or Caltech -the best way is to compare
it to the machine I’m typing on which is a classic computer.
The MacBook Air runs on an M1 chip that contains more than 16 billion transistors. Each
transistor can represent the “1” or “0” of binary data at the same time at least a little. The sheer
quantity of transistors makes the machine have computing capability.
A whopping 16 billion transistors are packed on the 120.5 square. A millimeter chip is an
enormous amount of transistors. TRAGIC, the first computer that was transistorized, had fewer
than 800. The ability of the semiconductor industry to add ever-growing transistors on a chip as
predicted by Intel Co-Founder Gordon Moore in the law named after him is what enabled the
exponential increase in computing power which has also created the basis for almost everything
There are certain things traditional computers aren’t able to do that quantum computers will
never be able of doing regardless of the number of transistors packed into an inch of silicon
inside the Taiwan semiconductor manufacturing facility (or “fab,” in the industry language).
That’s where the distinct and, frankly, bizarre characteristics of quantum computers are apart.
Quantum computers process information with qubits, which are able to be used to represent “0”
and “1” simultaneously. How do they achieve this? It’s a stretch for my expertise, but in essence,
qubits utilize the phenomenon of quantum mechanics, also known by the name of
“superposition,” whereby the properties of certain subatomic particles cannot be defined until
they are measured. Consider Schrodinger’s cat as at the same time alive and dead until you
unlock the box.
A single qubit can be cute however, things become thrilling when you begin adding more. The
computing power of traditional computers grows by linearly increasing the number of each
transistor, however, quantum computers’ power increases exponentially with each addition of
every new and reliable qubit. This is due to a quantum mechanical property known as
“entanglement,” whereby the individual probabilities for each qubit are affected by other qubits
within the system.
All of this implies that the limit of a quantum computer’s capabilities is far greater than the
capabilities of traditional computing.
Quantum computers can possibly solve issues that a traditional computer, however powerful,
could never solve. What kinds of issues? What is the nature of reality in the material realm that,
after all, is based on quantum mechanics and not classical mechanics? (Sorry, Newton.)
“Quantum computers simulate problems that we find in nature and in chemistry,” stated Jay
Gambetta, IBM’s vice director of quantum computing.
Quantum computers could model the physical properties of a hypothetical battery, allowing them
to design one that is much superior and more efficient than current versions. They can unravel
complex logistical issues, find the best routes to deliver goods, and improve predictions in
On the security front, quantum computers can breach cryptography protocols, rendering
everything from email information about national secrets to financial records unsecured. That’s
why the race to win quantum supremacy is an international contest and one that China and its
Chinese government has poured billions of dollars into. This has led the White House earlier
this month to publish a memo to establish a nation’s quantum computing leadership and to
prepare the nation for cybersecurity threats based on quantum technology.
Beyond the security concerns as well as the potential financial rewards could be huge.
Businesses are already offering quantum-computing-related services through the cloud to
customers such as ExxonMobil and the Spanish bank BBVA. The global market for quantum
computing was valued at under $500 million by 2020, International Data Corporation estimates
this market will be worth $8.6 billion in revenue by 2027 which will include more than $1.6 billion
But all of that won’t be feasible unless researchers are able to accomplish the engineering feat
of transforming the quantum computer which is a large scientific experiment into a commercially
The cold room
Within the Watson structure, Jerry Chow — the IBM’s director of IBM’s quantum computer center
opened a 9-foot-high transparent glass cube that appeared as a chandelier made of pure gold.
IBM’s Quantum System One. The chandelier is actually a high-tech refrigerator, equipped with
coils that contain superfluids that can cool the hardware to a hundredth of an inch Celsius over
absolute zero cooler, Chow told me, then space.
Refrigeration is the key ingredient to making IBM’s quantum computers run as well, and it’s also
the reason why this is an engineering problem. Although quantum computers could be
significantly more powerful than predecessors, they’re much more difficult to work with.
Remember my previous post about quantum properties like superposition and Entanglement?
Qubits are able to do things that even a tiny fraction of a pixel could not even imagine. Even the
slightest shift in temperature, noise or radiation can cause them to be stripped of their properties
by the process of decoherence.
This fancy refrigeration has been designed to stop the system’s qubits from decoding prior to
the computer has finished its computations. The earliest quantum qubits that were
superconducting lost their coherence under a nanosecond but today IBM’s top-of-the-line
quantum computers maintain coherence for up to 400 milliseconds. (Each second is comprised
of millions of microseconds.)
The problem IBM along with other corporations are facing is constructing quantum computer
systems that are not susceptible to error as well as “scaling the systems beyond thousands or
even tens of thousands of qubits to perhaps millions of them,” Chow explained.
This could be years away. In the year 2000, IBM introduced the Eagle 127-qubit processor and,
in its latest technical plan, the company plans to reveal a 433-qubit processor named the
Osprey in the coming months, and a 4,000-plus qubit machine in 2025. In 2025 quantum
computing may surpass the stage of experimentation, IBM CEO Arvind Krishna said to
journalists during a media conference earlier in the month.
A lot of experts are skeptical of whether IBM as well as any other of the rivals is ever going to
get there and raise concerns that some of the engineering issues that quantum computers pose
are too difficult for the system to ever be completely reliable. “What’s happened over the last
decade is that there have been a tremendous number of claims about the more immediate
things you can do with a quantum computer, like solve all these machine learning problems,”
Scott Aaronson, an expert in quantum computing in the University of Texas, said to me last year.
“But these claims are about 90 percent bullshit.” To meet that promise, “you’re going to need
some revolutionary development.”
In an increasingly digital age, advancement will depend on our ability to extract every bit of
value from the machines we design. It will all depend on the work of scientists who are like
Chow and his coworkers working in unlit labs to make innovation around some of the most
difficult computer engineering problems -and in the process build the next generation of