Risk Is Rising for a Quantum Winter for Quantum Computing in Two to Three Years

60 min readMar 9, 2022

Hype and expectations for quantum computing are growing much more rapidly than actual technical advances in the technology, so that patience may reach the breaking point in two to three years (2024 to 2025), leading to a quantum winter in sentiment, investment flows, budgets, and project approvals. It’s by no means a slam dunk that a quantum winter will ensue, but there simply is a rising risk as technology developers struggle to cope with outsize promises but limited high-end technology to match it. The solution? Trim premature efforts at commercialization but double down on basic research in the underlying hardware technology and algorithm and application techniques. This informal paper will explore the nature of the problem and how to circumvent it.

To be sure, no Quantum Winter is imminent in the coming months or year, but going out two to three years the risks are rising. It’s all a question of whether the pace of progress can be maintained and even accelerated. We can look forward to amazing strides of progress, but sustaining that more than two or three years could be a real challenge. Time will tell. I’m optimistic, but I’m a pragmatic realist as well.

Topics discussed in this paper:

In a nutshell — A Quantum Winter is unlikely in two to three years, but…
A Quantum Winter is unlikely in two to three years based on my personal expectations for progress, but… I’m an optimist, but… I’m also a pragmatic realist and the risks are rising
Definition for technological winter
Definition for Quantum Winter
Technological seasonal cycle: winters, springs, summers, and falls
Quantum winters, springs, summers, and falls
The life of a technology can have any number of cycles of winter, spring, summer, and fall
Quantum computing will have any number of cycles of winter, spring, summer, and fall
Seasonal cycles are typical for advanced technologies, so why should quantum computing be any different?
Definition for a technological summer
Definition for a quantum summer
Definition for a technological spring
Definition for a quantum spring
Technological falls
Quantum falls
Definition for a technological fall
Definition for a quantum fall
A technological winter is not the end of a technology but simply the end of one cycle of growth of the technology
A Quantum Winter is not the end of quantum computing but simply the end of one cycle of growth of quantum computing
A Quantum Winter is more of a pause or timeout than an ending
Stages of a Quantum Winter
A Quantum Winter is largely psychological in nature
Sentiment matters
Onset — How does a Quantum Winter start?
What might be the straw that breaks the camel’s back?
How might it get started?
A Quantum Winter is less than likely, but… risk is there and gradually growing
Yellow flags — Rising concerns about quantum computing
Red flag: Hype and expectations are growing faster than the technology itself
Still a mere laboratory curiosity
Still more appropriate for the lunatic fringe rather than mainstream application developers
How patient will investors and managers be if advances are not as rapid as expected?
Patience will be the key factor determining the onset of a quantum winter
Chief characteristics of a technological winter
What will it be like while it’s happening?
Telltale advance signs that a technological winter may be brewing
So far, so good, but…
Some telltale technical signs indicating rising risk of disappointment in quantum computing
Non-technical warning signs for quantum computing
Rising deficit of promises not yet fulfilled
My disappointment with the IBM 127-qubit Eagle
Attention quantum aficionados at large corporations — make sure your management and executive teams are well aware of these risks
Stagnation? Not so far
Will achieving The ENIAC Moment be critical to avoiding a Quantum Winter?
Really need to hit The ENIAC Moment before people can finally feel that quantum computing is real and not a vague and distant promised land
Is quantum error correction (QEC) needed to avert a Quantum Winter?
If we don’t have quantum error correction, then we do need near-perfect qubits to avoid a deep Quantum Winter
Will lack of higher qubit fidelity and enhanced qubit connectivity in IBM’s 433-qubit Osprey be enough to trigger a Quantum Winter?
What will Osprey deliver?
How far can we get without a high-level programming model and rich set of algorithmic building blocks?
Current technology has plenty of runway to avoid a Quantum Winter
Advances we need to see over the next two years to stay on track
Yes, we can indeed count on further progress, but will it be enough to keep us on track and to sustain momentum?
Sustaining momentum is everything — a slip in momentum can trigger a Quantum Winter
Will there be a quantum winter? Or even more than one?
But occasional Quantum Springs and Quantum Summers
Quantum Falls?
In truth, it will be a very long and very slow slog
No predicting the precise flow of progress, with advances and setbacks
Is a Quantum Winter likely in two to three years? No, but…
Critical technical gating factors which could presage a Quantum Winter in two to three years
Some technical walls could get hit
Will trapped-ion qubits and neutral-atom qubits save the day?
Will new and innovative qubit technologies appear on the scene and save the day?
Premature commercialization is probably the single biggest risk for stumbling into a Quantum Winter
Might we simply sleepwalk into a Quantum Winter?
Different audiences and sectors will experience a Quantum Winter differently or maybe even not at all
Risk of Quantum Winter is primarily for those engaged in premature commercialization — those involved with research and pre-commercialization should be fine
Quantum Ready — All dressed up and no place to go
We’re still in the full-on bliss of the honeymoon, but for how long?
Saving grace: Nobody is calling quantum a mania or a bubble, yet
What will assure that we can avoid this impending Quantum Winter?
Even minimal quantum advantage may remain out of reach in two to three years, but achieving it could avert a Quantum Winter
Single best way to avoid a Quantum Winter: Hold off on commercialization but double down on pre-commercialization
Focus on research
Wildcards
What single advance within three years could turn the tide?
Configurable packaged quantum solutions are the greatest opportunity for widespread adoption of quantum computing
Roadmaps would help avoid disappointment — if they are informative and reasonably accurate
My personal disappointments
No quantum applications yet — a great unrealized promise
General lack of scalable quantum algorithms and applications — they don’t seem to be a priority
Will Shor’s factoring algorithm be able to factor even 8 to 12-bit semiprime integers?
When might a Quantum Winter start?
How long might a Quantum Winter last?
It takes a protracted period of time to restore confidence
What ends a Quantum Winter?
Innovation to end a Quantum Winter
Why haven’t we seen a Quantum Winter yet even though past progress over the past 25 years was frequently quite slow?
My original proposal for this topic
Summary and conclusions

In a nutshell — A Quantum Winter is unlikely in two to three years, but…

A Quantum Winter is unlikely in two to three years based on my personal expectations for progress, but… I’m an optimist, but… I’m also a pragmatic realist and the risks are rising.
Quantum Winter is unlikely, but… Quantum computing is making great progress, but it is uneven and there are warning signs that if left unchecked could lead to a Quantum Winter in two to three years.
The tsunami is building slowly — but surely. Fueled by the raw energy of the massive hype and exaggerated and unrealistic expectations. Or large expectations which may be met, but might not be fully met. Maybe progress will be deep enough in two years that the energy of the tsunami will be distributed widely and not in some great wave that triggers a deep Quantum Winter which wipes out the field for many months or even years, or maybe it actually will be a wipeout. Flip a coin?
Patience will be the key factor determining the onset of a quantum winter. Those with great patience will not be causing — or experiencing — a Quantum Winter. Patient capital will prevail.
A Quantum Winter is largely psychological in nature. Sure, it’s based on issues with the technology, but it is the impact on human psychology that permits the technical issues to drive people into a Quantum Winter.
Sentiment matters — positive sentiment is needed. Too much negative sentiment — two to three years from now, as opposed to today — could easily trigger a Quantum Winter. Too much neutral or indifferent sentiment could cause a Quantum Winter as well.
If we do see a Quantum Winter it will be after we spend two years to presumably get to being on track for production-scale quantum applications, and then one year for disappointment and disenchantment to unfold into deep depression if it turns out that we are not on track after two years from now.
Nominally in two years if not enough progress. Need near-perfect qubits — four or 3.5 nines. Need closer to full connectivity. Need coherence sufficient for fairly deep circuits. Need fine granularity of phase for 20 to 40-qubit quantum Fourier transform (QFT) and quantum phase estimation (QPE).
Could start earlier than two years. If insufficient progress and rising disenchantment and disillusionment.
Could start later than two years. If reasonable progress, but a few lingering key items and it just takes another few months or a year to realize that those items are just not as imminent as promised.
People are starting to drink way too much of the marketing Kool-Aid. The elixir of impending commercialization is beginning to take root. And people are reluctant to commit to the hard work and slow slog of pre-commercialization, especially calm and gradual research.
Still a mere laboratory curiosity. Quantum computing is still at the stage of being a mere laboratory curiosity, not even close to being ready for development and deployment of production-scale practical real-world quantum applications.
Still more appropriate for the lunatic fringe rather than mainstream application developers.
Premature commercialization. Single best way to cause a Quantum Winter is to pursue premature commercialization before the technology actually is ready to develop and deploy production-scale practical real-world quantum applications.
Best way to avoid Quantum Winter is to hold off on commercialization and double down on pre-commercialization. Single best way to avoid a Quantum Winter is to avoid premature commercialization and double down on per-commercialization instead, particularly research. Focus on research, prototyping, and experimentation. It’s not yet time to get commercial efforts underway.

A Quantum Winter is unlikely in two to three years based on my personal expectations for progress, but… I’m an optimist, but… I’m also a pragmatic realist and the risks are rising

I don’t have any difficulty seeing viable paths to avoiding a Quantum Winter in two to three years, but… that’s me speaking as an optimist. But as a pragmatic realist I have to admit that substantial technical risks abound and are rising.

Yes, there are plenty of technical obstacles to achieving production-scale practical real-world quantum applications in two to three years, but one should never underestimate the cleverness of engineers when faced with the most challenging problems.

But with so many technical challenges out there, and more arising every day, a pragmatist does have to wonder if we really can stay ahead of all of them.

Definition for technological winter

A Quantum Winter is simply a special case of the broader concept of a technological winter.

A basic definition is in order for the generic concept of a technological winter.

A technological winter is a period of stagnation in the progress of a technology.
Alternatively, a technological winter is a period of reduced funding and interest in a technology.

That’s just the raw definition. More details will be elaborated shortly.

Definition for Quantum Winter

A Quantum Winter is of course a specific type of technological winter:

A Quantum Winter is a period of stagnation in the progress of quantum computing.
Alternatively, a Quantum Winter is a period of reduced funding and interest in quantum computing.

The full details of a technological winter apply equally to a Quantum Winter as to a winter of any other form of technology. Those details will be elaborated shortly.

Technological seasonal cycle: winters, springs, summers, and falls

Technological advance can be seen as a cycle with seasons — winter, spring, summer, and fall.

The complement of a technological winter is a technological summer whose arrival is presaged by a technological spring.

Similarly, the complement of a technological spring is a technological fall which follows a technological summer and presages a technological winter.

Quantum winters, springs, summers, and falls

As with a technological winter, the complement of a quantum winter is a quantum summer whose arrival is presaged by a quantum spring.

Similarly, the complement of a quantum spring is a quantum fall which follows a quantum summer and presages a quantum winter.

The life of a technology can have any number of cycles of winter, spring, summer, and fall

A technology can have any number of winters, springs, summers, and falls over its life.

The durations and intensities of the cycles and seasons can vary greatly.

Quantum computing will have any number of cycles of winter, spring, summer, and fall

Quantum computing will have any number of winters, springs, summers, and falls over its life.

The durations and intensities of the cycles and seasons will vary greatly.

Seasonal cycles are typical for advanced technologies, so why should quantum computing be any different?

I’ve lost track of how many AI Winters artificial intelligence has been through — let alone how many more it may have. Many or most advanced technologies inevitably experience seasonal cycles of some sort, with one or even more technological winters, so it may be unrealistic to expect that quantum computing won’t have its own share of seasonal cycles.

Maybe the only real question here is whether quantum computing will make it to its initial widespread commercial success before slipping into a Quantum Winter.

I expect that quantum computing will have a progression of stages of commercial success, and that each stage has the risk of its own Quantum Winter.

Definition for a technological summer

A technological summer is a period when a technology flourishes with great success.
Alternatively, a technological summer is a period of increased funding and interest in a technology.

In other words, a technological summer is the opposite of a technological winter.

Definition for a quantum summer

A quantum summer is a period when quantum computing flourishes with great success, presaged by a quantum spring and presaging a quantum fall.
Alternatively, a quantum summer is a period of increased funding and interest in quantum computing.

In other words, a quantum summer is the opposite of a quantum winter.

Definition for a technological spring

A technological spring is a period of birth or renewal, when a technology begins to flourish with success, typically presaging a technological summer, and likely or possibly presaged by a technological winter.
Alternatively, a technological spring is a period of increasing funding and interest in a technology, typically presaging a technological summer, and likely or possibly presaged by a technological winter.

Definition for a quantum spring

A quantum spring is a period of birth or renewal, when quantum computing begins to flourish with success, typically presaging a quantum summer, and likely or possibly presaged by a quantum winter.
Alternatively, a quantum spring is a period of increasing funding and interest in quantum computing, typically presaging a quantum summer, and likely or possibly presaged by a quantum winter.

Technological falls

The complement of a technological spring is a technological fall which is presaged by a technological summer and presages a technological winter.

Quantum falls

The complement of a quantum spring is a quantum fall which is presaged by a quantum summer and presages a quantum winter.

Definition for a technological fall

A technological fall is a period of gradual decline after a technological summer, when a technology peaks, loses its luster, and begins to decline, presaging a technological winter.
Alternatively, a technological fall is a period of decreasing funding and interest in a technology after a technological summer, presaging a technological winter.

Definition for a quantum fall

A quantum fall is a period of gradual decline after a quantum summer, when quantum computing peaks, loses its luster, and begins to decline, presaging a quantum winter.
Alternatively, a quantum fall is a period of decreasing funding and interest in quantum computing after a quantum summer, presaging a quantum winter.

A technological winter is not the end of a technology but simply the end of one cycle of growth of the technology

A technological winter will not signal that a technology is dead for the rest of time, but simply indicates that one growth cycle has run its course.

A next cycle of growth is usually inevitable, but its precise timing is indeterminate. It could take months, years, or even decades for the next growth cycle to commence with a technological spring.

A Quantum Winter is not the end of quantum computing but simply the end of one cycle of growth of quantum computing

A Quantum Winter will not signal that quantum computing is dead for the rest of time, but simply indicates that one growth cycle for quantum computing has run its course.

A next cycle of growth for quantum computing is usually inevitable, but its precise timing is indeterminate. It could take months, years, or even decades for the next growth cycle to commence with a quantum spring.

A Quantum Winter is more of a pause or timeout than an ending

Just to reiterate the message from the preceding section, a Quantum Winter is not the end of the line for quantum computing, but a temporary pause or timeout, an opportunity to regroup and rethink approaches, before a new phase of growth, a quantum spring, a renewal, can begin anew.

Stages of a Quantum Winter

A Quantum Winter will proceed as a sequence of stages. The overall stages of a Quantum Winter include:

How does it actually start, the first moment? Such as trigger events.
What keeps it going? Lack of reasonable progress.
How long might it last? Months, years, or…?
What ends it? Such as trigger events of innovation or renewed flows of money.
What’s next? Quantum Spring of the next stage of growth.

A Quantum Winter is largely psychological in nature

Sure, a Quantum Winter is based on issues with the technology, but it is the impact on human psychology that permits the technical issues to drive people into a Quantum Winter. So a Quantum Winter is largely psychological in nature.

A Quantum Winter may unfold as a sequence of psychological stages:

Disappointment.
Disenchantment.
Disillusionment.
Despair.
Depression.

The psychological impact of technical tissues can start with relatively minor disappointments and gradually snowball downwards from there, getting worse with the passage of time.

Sentiment matters — positive sentiment is needed

Sentiment is a simple umbrella term for all aspects of the psychological nature of how people and organizations think and feel about a technology and its prospects. Generally, positive sentiment is needed to drive and sustain an endeavor.

Overall, sentiment can be:

Positive.
Neutral.
Negative.

If generally positive, sentiment can be:

Open-minded.
Positive.
Interested.
Optimistic.
Accepting.
Committed.
Buoyant.
Enthusiastic.
Energized.
Ecstatic.
Euphoric.

If generally neutral, sentiment can be:

Undecided.
Neutral.
Indifferent.
Disinterested.
Skeptical.
Open-minded.
Noncommittal.

If generally negative, sentiment can be:

Skeptical.
Pessimistic.
Cynical.
Opposed.
Anxious.
Disappointed.
Disenchanted.
Disillusioned.
Depressed.

A Quantum Winter is unlikely when sentiment is generally positive.

A Quantum Winter is less likely but still possible when sentiment is generally neutral. Lack of enthusiasm can drain the energy from an endeavor.

A Quantum Winter is very likely when sentiment is generally negative. Without generally positive energy, there is little to drive and sustain momentum.

Onset — How does a Quantum Winter start?

A Quantum Winter will be preceded and presaged by a quantum fall, which is similar to a Quantum Winter except much milder — a modest to moderate slowdown in sentiment, interest, activity, and the flow of money.

Some indicators to watch for:

Trigger events. Just feels that something isn’t quite right. Bad events, or failure of good events to happen.
Warning signs. Telltale advance signs that something bad is going to happen. People are just acting a bit odd and out of character.
How it feels. A visceral sense that something is’t right and that bad things are beginning to happen.
Consequences. The effects of trigger events.
Appearance. What it looks like — and something just doesn’t look right.

What might be the straw that breaks the camel’s back?

There could be a wide range of trigger events for a Quantum Winter, such as:

Some high-profile commitment to achieve some grand goal by some particular date — and then failing to achieve it. Such as heavy PR for a failed attempt at The ENIAC Moment.
Bankruptcy or shutdown of a high-flying startup or corporate joint venture.
Some high-flying startup fails to close a massive funding round.
Failure of some large government-sponsored project.
Qubit fidelity could be only so-so or actually okay, but transmon qubit fidelity may still be very weak. And trapped-ions may have great connectivity, but still mediocre qubit counts.
Inability to achieve nontrivial quantum Fourier transform (QFT) or quantum phase estimation (QPE).

Those are just some examples. It could be anything that causes sentiment to head south.

How might it get started?

Here’s one scenario (among many) for how a Quantum Winter could get started:

Enough people have bought into the euphoria and gotten convinced to expect real results within six months.
If a critical mass of people hit that six month milestone and are very disappointed in mediocre or even nonexistent results, that could trigger the beginning of the slide down the slippery slope to a Quantum Winter.
Rinse and repeat.

A Quantum Winter is less than likely, but… risk is there and gradually growing

I honestly don’t expect that a Quantum Winter will actually occur two to three years from now, but I can’t completely discount it either. There is risk and it’s growing, albeit gradually. There are just too many yellow flags to ignore, both for hardware and algorithms and applications.

Yellow flags — Rising concerns about quantum computing

Some of the yellow flags I see that could contribute to a Quantum Winter:

Despite dramatic advances, still not close to commercial readiness.
Still a mere laboratory curiosity. Quantum computing is still at the stage of being a mere laboratory curiosity, not even close to being ready for development and deployment of production-scale practical real-world quantum applications.
Still more appropriate for the lunatic fringe rather than mainstream developers.
Irrational exuberance and hype. High and rising.
Hype and expectations are growing faster than the technology itself.
Technical concerns not being addressed rapidly enough.
Lots of qubits, but not so usable.
Technical risks are quite daunting.
Key and essential technical hurdles don’t have clear and compelling roadmaps.
Rush to premature commercialization.
Too much money chasing two few real opportunities.
Too many research questions remain outstanding.
The technology just doesn’t feel ready for commercial use despite the advances.
Dearth of 40-qubit algorithms.
No sense of real and substantial quantum advantage.

Oh, but other than that, how was the play, Mrs. Lincoln?!!

Red flag: Hype and expectations are growing faster than the technology itself

I think my biggest concern relative to the risk of a Quantum Winter is the simple fact that the hype and expectations are growing faster than the technology itself. Granted, that could change at any moment if the pace of advance of the technology accelerates, but that’s not a slam dunk since the technology is very difficult, so the pace of advance has remained quite modest even if it nonetheless is quite impressive.

Still a mere laboratory curiosity

Quantum computing is still at the stage of being a mere laboratory curiosity, not even close to being ready for deployment of production-scale practical real-world quantum applications.

Much research is still required. Many technical issues remain to be resolved.

Granted, as a laboratory curiosity it is indeed quite appropriate to prototype systems and to experiment with quantum algorithms and quantum applications.

But prototyping and experimentation should not be confused with product engineering and development and deployment of production-scale practical real-world quantum applications.

Being a mere laboratory curiosity is fine for where we are today, focused on prototyping and experimentation, but we run the risk of slipping into a Quantum Winter if we’re still at this stage of being a mere laboratory curiosity two to three years from now.

For more discussion of quantum computing being a mere laboratory curiosity, see my paper:

Still more appropriate for the lunatic fringe rather than mainstream application developers

The lunatic fringe are those super-elite technical staff who are capable and interested in working with a new technology regardless of whether the technology is ready for commercial deployment. Quantum computing is still at the stage where its primary appeal is to the lunatic fringe rather than to mainstream application developers.

This is okay for where we are today, but two to three years from now it will be necessary to cater to mainstream application developers rather than the lunatic fringe.

Quantum computing runs the rising risk of falling into a Quantum Winter if it still only appeals to the lunatic fringe two to three years from now.

For more on the lunatic fringe, see my paper:

How patient will investors and managers be if advances are not as rapid as expected?

Sure, the early, more adventurous investors and managers are likely to have a substantial level of strategic patience, but the vast majority of newcomers, late arrivals at the party, are not so likely to be so patient.

How patient will investors, managers, executives, and boards be as small initial experimental teams grow to larger budgets but fail to deliver production-scale applications within two or three years?

Increasingly high levels of investment and management attention will require a high rate of return and turn quickly into severe disappointment if high returns are not achieved in short order.

Some potential consequences if progress stalls or is even modestly delayed:

Projects could get canceled.
Budgets could be slashed.
Hiring could be curtailed.
Staff could be cut.
Investment flows could dwindle.

But research budgets should be safe, being focused on a longer-term horizon.

Premature decisions to focus on commercialization rather than research, prototyping, and experimentation — collectively referred to as pre-commercialization — will tend to prove to have been unwarranted.

Technical staff and management should stay focused on pre-commercialization in order to avoid getting sucked into a deep Quantum Winter.

Patience will be the key factor determining the onset of a quantum winter

Those with great patience will not be causing a Quantum Winter. Patient capital will prevail.

Rather, those without any significant patience will be the cause of a Quantum Winter. Impatient capital — will lead to capital flight. These are the people who are unable to persist and wait for long-term developments to play out.

When it comes to budgets, investments, and projects, it will come down to a question of who has a long-term view (patient capital) as opposed to those who have a short-term view (impatient capital.)

Chief characteristics of a technological winter

A Quantum Winter will have the same general characteristics as a winter for any technology or innovation.

The chief characteristics of a Quantum Winter or a technological winter in general will be:

Disappointment with the technology. Lost some or all of its luster.
Failure of the technology to deliver on its promises.
Rising prospect or concern that the technology will fail to deliver on its promises in the future. The future no longer feels as bright or even bright at all.
When predictions fail to materialize.
Stagnation. Lack of palpable progress or technical advances. It needs to feel and look like the the technology is progressing at some reasonable pace.
Over-investment and over-enthusiasm.
Rising prospect that projects will fail to deliver on their promises.
Expectations feel unfulfilled.
Feeling a lack of progress. Or feeling of dramatically less progress than expectations,
Diminished enthusiasm and excitement.
Rising frustrations.
Intense frustrations.
Essential technical capabilities not achieved.
Key technical milestones not achieved.
Getting bogged down with technical speed bumps. Can see a path forward but unable to readily achieve it.
Hitting technical walls. No clear path forward.
Progress has halted. Any number of causes.
Progress has dramatically diminished. Any number of causes.
Progress is haphazard. Uneven and not meeting expectations.
Fewer articles and papers published.
Progress without results. Sense of at least some progress in terms of expenditure of effort and money but with a lack of palpable and visible results which deliver significant business value to the organization.
Anxiety. General feeling that things could get worse. Feeling that things are beginning to fall apart.
Despair and even depression. General feeling that things are not going to get better. Feeling that things are likely to get much worse.
Curtailment of projects. Existing projects are limited or trimmed, or even canceled. New projects are not approved or funded, or receive dramatically less funding.
Less money readily available.
Curtailment of money flows. Less investment in startups. Limited or canceled budgets.
Limitations on hiring. Due to curtailment of projects and money flows. Fewer open positions. Less easy to find work. Less easy to hop between companies.
Trimming of research funding. Although generally research will be impacted less than commercial projects — unless things are really bad.
General collapse of the field. Insufficient demand for so many vendors, customers, projects, staff, initiatives, office space, furnishings, equipment, budgeting, and investment.

What will it be like while it’s happening?

All or most current hardware will still be operational. Some may get shut down.
Less frequent advances.
More inverse hype — disappointment and despair. The opposite of euphoria.
Less money sloshing around. And less enthusiastically.
Fewer job opportunities.
Fewer new projects.
Fewer new startups.

Telltale advance signs that a technological winter may be brewing

All of these telltale signs of a possible impending technological winter are generic and apply to all technologies, but certainly to quantum computing in particular:

Rampant hype.
Over-inflated promises.
Unnaturally high expectations.
Excessive hype coupled with nonexistent practical solutions.
Excessive promotion by the media.
Enthusiasm and optimism for the technology has spiraled out of control.
Irrational exuberance and hype. High and rising, faster and faster.
Enthusiasm for the promise of the technology has spiraled out of control. Even before the technology is actually available.
Rate of progress is failing to meet expectations.
Despite dramatic advances, still not close to commercial readiness.
Technical concerns not being addressed rapidly enough.
Technical risks are quite daunting.
Key and essential technical hurdles don’t have clear and compelling roadmaps.
Rush to premature commercialization.
Too much money chasing two few real opportunities.
Too many research questions remain outstanding.
The technology just doesn’t feel ready for commercial use despite the advances.

So far, so good, but…

Quantum computing has made a lot of excellent technical progress over the past few years, but it has been uneven and with gaps, limitations, and issues.

I remain hopeful for the next two to three years, but… reality remains to be seen. Or I should say that the promises remain to be seen — turned into reality.

Plenty of technical advances are near-certain slam dunks, but… plenty of the technical hurdles may be problematic.

Some telltale technical signs indicating rising risk of disappointment in quantum computing

There are lots of telltale warning signs in quantum computing already, now.

It will be difficult to believe that quantum computing is on track to a bright future if these technical hurdles still remain in two to three years:

No 40-qubit algorithms. No algorithms approaching production-scale that can be run on simulators even if real quantum hardware is not yet available.
Quantum volume stuck below 16M (24 qubits). Not even close to 16 bits or 20 bits.
Few 24-qubit algorithms. Currently none. Max at 21 to 23 qubits.
Mediocre qubit fidelity. Low qubit fidelity. Qubit fidelity stuck below 3.5 nines (99.95).
Quantum error correction (QEC) is a distant future. The promised land of perfect logical qubits, but not in the near future. Too far in the future for most people to relate to it.
Near-perfect qubits are still too far in the future. A fair chance that we will see them in two to three years, but a risk that we won’t see them until it’s too late to prevent confidence from falling off a cliff into a Quantum Winter. But, if we see them within a year to 18 months, a Quantum Winter can be averted, assuming other factors are also addressed in an equally timely manner.
Weak connectivity for transmon qubits. Stuck with nearest neighbor connectivity. Although trapped-ion qubits have full any to any connectivity.
Lack of fine granularity of phase and probability amplitude. Needed to enable quantum Fourier transform (QFT) and quantum phase estimation (QPE) for nontrivial use cases, especially for quantum computational chemistry.
No common use of quantum Fourier transform (QFT) or quantum phase estimation (QPE) for nontrivial use cases. Nothing using 16 or 20 or 24 or 28 or 32 or 40 bits, let alone 64 or 80 or 96 bits.
No significant quantum advantage. Not even minimal quantum advantage (1,000X classical solutions), let alone substantial quantum advantage (1,000,000X), and definitely not close to dramatic quantum advantage (one quadrillion X.)
No sense that The ENIAC moment is coming any time soon. Nothing even remotely close to a production-scale application.
No rich set of algorithmic building blocks. Makes it very difficult to construct large, complex, and more sophisticated algorithms.
No high-level programming model. Extremely low-level programming model. How far can we get with the current low-level programming model? Or as I used to say back in the 1970’s, how much can you expect people to do with bear skins and stone knives.
General lack of scalable quantum algorithms. Doesn’t seem to be a priority.
No quantum applications yet. A great promise for the future, but not realized in any way yet.
Mediocre and limited simulators. Classical simulation of quantum circuits is a very powerful tool. We could be doing much better, especially performance, capacity, and debugging and analysis tools.
Weak debugging capabilities and analysis tools. We are getting by since our algorithms are currently quite small and simple. Much larger, complex, and more sophisticated algorithms will require dramatically greater debugging and analysis tools.
Disappointment with IBM 127-qubit Eagle. Plenty of qubits, but only a quantum volume of 64. Mediocre qubit fidelity. Weak qubit connectivity.
Low confidence in the IBM 433-qubit Osprey due later this year. The disappointment in Eagle makes it difficult to have much confidence in Osprey. This is exactly the kind of falling confidence that can bring on or accelerate a Quantum Winter.
Rigetti just introduced 40 and 80-qubit processors, but with similar limitations as IBM Eagle. Four years ago they promised 128 qubits within a year.
IonQ is making sluggish and haphazard progress. Not clear when they will have a production-scale machine, capable of production-scale 40 to 80-qubit algorithms and applications.
Honeywell (now Quantinuum) made a big splash but with too few qubits. They did promise to improve quantum volume by an order of magnitude each year, but that’s only an average of 3.25 additional qubits each year (log2(10) = 3.32) — with 12 qubits now, that would be 18 to 20 qubits in two years and maybe 22 qubits in three years, which is not so inspiring of the great confidence which will be needed to avoid a Quantum Winter.
Still too many new papers with algorithms using only five to eight qubits. Where are the 40-qubit algorithms — which can be simulated even if high-fidelity real quantum hardware is not yet available?

Non-technical warning signs for quantum computing

Excessive hype coupled with nonexistent practical solutions.
Plenty of decent advances in recent years, but the pace of growth of hype and expectations exceed the pace of technical advances.
Over-investment and over-enthusiasm. Feels like we might be on the cusp of that over the next 18 months.

Rising deficit of promises not yet fulfilled

A major concern with quantum computing is that so many promises are being made but so few of them have come to fruition and success, so far. Each of these promises is a liability or deficit, which if not paid in the not too distant future will contribute to sending quantum computing into a Quantum Winter. The deficit is rising rapidly, with no end in sight.

Where are all of the 40-qubit algorithms?

It is very telling that there is a dearth of 40-qubit algorithms at this time. Sure, we currently lack real quantum hardware capable of supporting 40-qubit algorithms with high fidelity, but we do have classical simulators which can, so there isn’t a lot of great excuse for there being virtually nothing in terms of algorithms approaching at least a fraction of production-scale that can be run on simulators even if real quantum hardware is not yet available.

But as quantum hardware becomes more capable over the next two to three years, I would hope that we should start seeing 32 to 40-qubit algorithms.

The point is that if we aren’t seeing a plethora of 40-qubit algorithms (and scalable 40-qubit algorithms at that) two to three years from now, it won’t bode well for avoiding a Quantum Winter.

In short, if we don’t have 40-qubit algorithms in two to three years, then we don’t have anything — except a lot of toy algorithms.

For more on this issue of 40-qubit algorithms, see my paper:

My disappointment with the IBM 127-qubit Eagle

As already mentioned, I’m rather disappointed with the IBM 127-qubit Eagle quantum processor. It’s brand new (November 2021), so it should be the best anybody can do, but it’s rather disappointing. It has plenty of qubits, but only a quantum volume of 64 — only six qubits can be used at a time for a high-fidelity quantum computation. It has mediocre qubit fidelity and weak qubit connectivity. And no hint of a specification of what granularity of phase and probability amplitude it supports.

In short, it doesn’t appear to be on track to avoid a Quantum Winter in two to three years.

For a lot more detail on my critique of Eagle, see my paper:

Attention quantum aficionados at large corporations — make sure your management and executive teams are well aware of these risks

Best defense from getting slammed by a Quantum Winter is to be preemptive.
Best if management hears this from you first. The best defense is a strong offense. Once you’re on the defensive, you lose.
Lower expectations. Try really hard to set lower expectations and focus on research, prototyping, and experimentation — pre-commercialization — rather than prematurely pushing for commercialization.

Stagnation? Not so far

One telltale sign of a technological winter is stagnation — the lack of progress or technical advances, but so far, quantum computing is making a reasonable sense of progress, with only very brief periods of calm and waiting.

Generally, we’re in a period where virtually every day brings a veritable blizzard of fresh announcements — technical, commercial, and financial — and publication of papers.

So by at least this one metric, quantum computing is doing great and unlikely to fall into a deep Quantum Winter any time soon.

Will achieving The ENIAC Moment be critical to avoiding a Quantum Winter?

The ENIAC Moment will be the first time that a production-scale practical real-world quantum application can be demonstrated on a real quantum computer. Whether that breakthrough milestone can be achieved in two to three years remains to be seen. It’s very possible, even possibly likely, but not a slam-dunk certainty. The question here is whether achieving that milestone is mandatory to avoiding a Quantum Winter.

I would say that it is not 100% mandatory, but we do need to be close enough so that confidence in the technology is not damaged so badly that slipping into a Quantum Winter is unavoidable.

Carefully and artfully managing expectations can indeed avoid a Quantum Winter, but that’s the ultimate risk for technological winters — that expectations have been managed poorly and zoom far ahead of reality, sending the field sliding into a technological winter before you even know it and without even realizing that it’s happening.

Let’s hope that expectations for The ENIAC Moment can be managed in a pragmatic manner.

For more on The ENIAC Moment, see my paper:

Really need to hit The ENIAC Moment before people can finally feel that quantum computing is real and not a vague and distant promised land

Although I wouldn’t say that achieving The ENIAC Moment is absolutely essential for avoiding a Quantum Winter in two to three years, I would say that it is critically urgent and we really do need to achieve it as soon as possible. If not in two to three years, then certainly within three to four years.

So even if we haven’t slid into a deep Quantum Winter in two to three years, a general sense of modest to moderate disappointment would be a real drag on further progress.

If anything, in two to three years we need to see a clear gathering of momentum towards The ENIAC Moment within another year or two years or so. It at least needs to look and feel close and palpable.

Is quantum error correction (QEC) needed to avert a Quantum Winter?

Perfect logical qubits are essential for the long-term health and viability of quantum computing, but my conjecture is that most quantum algorithms and quantum applications will be able to get by with near-perfect qubits at least for the next few years rather than require perfect logical quits which require full, automatic, and transparent quantum error correction (QEC).

So, I don’t see the lack of quantum error correction as sending the field into a deep Quantum Winter. But, I could be wrong about that.

Actually a lack of quantum error correction and a lack of near-perfect qubits would probably be sufficient to send us spiraling into a deep Quantum Winter. We would need qubit fidelity of at least three nines or possibly 3.5 nines to avoid a Quantum Winter.

So, as long as we do have near-perfect qubits we don’t need quantum error correction to avoid a deep Quantum Winter.

That said, a delay in achieving perfect logical qubits further down the road, say in five to seven years, might in fact trigger a deep Quantum Winter. It all depends on so many factors.

Quantum error correction is likely to take more than another three years to achieve practical and economical implementations — maybe in four years, possibly not for five years, well beyond the period of concern for this paper.

Anybody who is falsely led to believe that quantum error correction would be ready in less than three years will likely be very disappointed, very disenchanted, very disillusioned, and more likely to jump ship and abandon quantum computing once a Quantum Winter begins. So for those people, the lack of quantum error correction could indeed trigger a Quantum Winter in the next two to three years.

Those who are led to believe that quantum error correction will be available in a year to 18 months will be the vanguard of a Quantum Winter.

We do need to carefully distinguish those who could use quantum error correction from those who critically require it. The former could be more patient while the latter will not only be impatient, but more likely to drop quantum computing like a hot potato when it finally sinks in that quantum error correction won’t be practical in the next two to three years.

IBM has explicitly promised that each new machine will come one step closer to supporting quantum error correction — “every processor we design has fault tolerance considerations taken into account.” IBM hasn’t made any explicit and specific promises about the 433-qubit Osprey which is due out later this year (2022), but given the large number of qubits — which no existing algorithms would be ready to utilize, they will need to demonstrate five to eight logical qubits with quantum error correction and six nines of qubit fidelity for logical qubits later this year to maintain any sense of credibility. Failure to do so will send shock waves though the field. It may not be enough to trigger a full Quantum Winter by itself, but it would be enough to put everyone on edge so that even some minor additional setbacks, failed advances, or delayed advances could be the straw that breaks the camel’s back and kicks off a deep Quantum Winter.

For more on quantum error correction, logical qubits, and fault-tolerant quantum computing, see my paper:

If we don’t have quantum error correction, then we do need near-perfect qubits to avoid a deep Quantum Winter

Just to reemphasize something buried in the preceding section, a lack of quantum error correction and a lack of near-perfect qubits would probably be sufficient to send us spiraling into a deep Quantum Winter. We would need qubit fidelity of at least three nines or possibly 3.5 nines to avoid a Quantum Winter.

So, as long as we do have near-perfect qubits we don’t need quantum error correction to avoid a deep Quantum Winter.

But without either, we would definitely be likely to spiral down deep into a Quantum Winter two to three years from now.

We can continue to limp along for the next year or even 18 months, but only based on expectations that near-perfect qubits will be available shortly — within two to three years from now.

Will lack of higher qubit fidelity and enhanced qubit connectivity in IBM’s 433-qubit Osprey be enough to trigger a Quantum Winter?

Regardless of whether Osprey supports quantum error correction (QEC) and perfect logical qubits, the pressure is on for IBM to deliver higher qubit fidelity and greater qubit connectivity than they have for the 27-qubit Falcon and the 127-qubit Eagle quantum processors. Failure of IBM alone to deliver such advances may not be sufficient to trigger a deep Quantum Winter, but it is still possible, depending on whatever else is going on. It might just trigger a shift from transmon qubits to trapped-ion or neutral-atom qubits and send transmon qubits alone into their own special Quantum Winter.

It’s possible that Osprey may need to demonstrate qubit connectivity improvements beyond nearest neighbor connectivity in order to not completely lead to a loss of confidence, and possibly lead to the kick-off of a quantum winter. Literally, anything is possible when people are very disappointed when expectations are set too high.

All of that said, Osprey is due out in under a year, so that would nominally be too soon for a true Quantum Winter per se. But, it could indeed lay the foundation for enough lack of confidence that failure to deliver improvements over the subsequent year, when the 1,121-qubit Condor processor is due from IBM could well finally deliver the onset of the feared Quantum Winter.

That said, IBM is not the only game in town, so failures on IBM’s part are not fatal alone, but they certainly can set the stage for the onset of a Quantum Winter if there are enough other disappointments as well.

What will Osprey deliver?

Right now, the promised IBM 433-qubt Osprey quantum processor is the biggest anticipated technical advance over the next year. The big question is whether it will be a relatively minor incremental advance or a major quantum leap.

Will Osprey be simply a next stage beyond the 127-qubit Eagle, but still just more noisy NISQ qubits, or will it be a big leap beyond NISQ qubits? We just don’t know right now. All we can do is hope and speculate.

There are three main worrisome fronts for Osprey:

Qubit fidelity. Will it be high enough?
Qubit connectivity. Will it be great enough?
Fine granularity of phase and probability amplitude. How large a quantum Fourier transform (QFT) or quantum phase estimation (QPE) will be supported?

Coherence time and circuit depth are looming issues as well, but are not front and center until the big three are resolved.

How far can we get without a high-level programming model and rich set of algorithmic building blocks?

The current programming model for quantum computing is rather low level, more like a classical assembly language or machine language than a modern high-level language. Similarly, we lack a rich set of high-level algorithmic building blocks for constructing complex and sophisticated quantum algorithms and quantum applications. Will it be enough to get us to where we need to be in two to three years to be on track to developing production-scale practical real-world quantum applications?

I surmise that super-elite technical teams will eventually be able to achieve The ENIAC Moment — the first production-scale practical real-world quantum application without the need for a high-level programming model and rich set of algorithmic building blocks, but most people (non-elites) will in fact need a high-level programming model and rich set of algorithmic building blocks.

That still leaves open exactly how far we can get using the kind of super-elite technical teams who don’t need a high-level programming model and rich set of algorithmic building blocks.

It also leaves open the question of whether we will likely stumble into a deep Quantum Winter if we don’t have access to a high-level programming model and rich set of algorithmic building blocks.

It’s hard to say, but my tentative conclusion is that we should be able to make it through the next two to three years using only super-elite technical teams who don’t need a high-level programming model and rich set of algorithmic building blocks.

But… for the rest of the people, those technical teams which are not super-elite, they may in fact stumble into a deep Quantum Winter even as the super-elite technical teams soldier on and manage to avoid stumbling into a Quantum Winter.

Current technology has plenty of runway to avoid a Quantum Winter

Although there are plenty of technical concerns, we still have plenty of time and plenty of technical opportunities (plenty of runway) sufficient to overcome the concerns over the next two to three years.

So there’s no need to worry per se.

But there is good reason to be concerned — so that issues can get the attention they need to avoid sliding into a Quantum Winter two to three years from now,

Advances we need to see over the next two years to stay on track

In a previous informal paper I enumerated the many advances that we need to see in quantum computing over the next two years or so to stay on track — for people to feel that we’re making great or at least decent progress towards making the design, development, and deployment of production-scale practical real-world quantum applications a reality not too many years after that. Failure to achieve some sort of critical mass of those advances would put us at risk of falling into a Quantum Winter.

Read my informal paper on these needed advances:

Yes, we can indeed count on further progress, but will it be enough to keep us on track and to sustain momentum?

I just don’t know the answer here, yet. Yes, it’s a slam dunk that we will see a lot of amazing progress over the next two to three years, but will it be enough to get close to supporting production-scale practical real-world applications on real quantum hardware, especially in the face of excessive hype and unreasonable and irrational expectations?

Momentum is a tricky thing. One minute you can have it and it’s great, but the next minute it may be gone. Sustaining it can be a very heavy burden.

We may see two years of great momentum, but then hit some walls or major speed bumps and then… sky-high expectations could undermine confidence overnight.

Again, there is no solid reason right now to presume a great risk in two to three years, but we need to keep our eye on the ball of managing expectations every step of the way for the next three to five years.

Sustaining momentum is everything — a slip in momentum can trigger a Quantum Winter

I can’t overestimate how important maintaining momentum will be over the next two to three years. We could get by with mediocre and spotty momentum over the past three to five to ten years since there was only minimal hype, minimal expectations, and minimal investment flows to deal with. But now, and accelerating over the next two to three years, we will face monumental rising hype and irrational expectations coupled with monumental capital investment flows, including corporate budgets, so that even relatively minor slips in momentum could present significant risk to damaging confidence in momentum, dramatically raising the risk of sliding into a Quantum Winter.

A few little slips in momentum here and there won’t likely be fatal, but any sense of accumulation of slips in momentum could well be fatal and trigger a severe Quantum Winter.

Will there be a quantum winter? Or even more than one?

Technological progress is really hard to predict. Sometimes progress is rapid, sometimes not so much. Sometimes there’s a long period of progress, and sometimes long periods of seeming lack of progress.

Quantum computing has made great progress in recent years, and much more progress is expected in the coming years, but breakthroughs are hard to predict on a schedule. And sometimes the most promising technologies just don’t work out.

There are several areas were dramatic advances are needed which may take much longer than expected:

Near-perfect qubits. High enough qubit fidelity — low enough error rate — that many applications don’t even need quantum error correction (QEC). And that also enables efficient quantum error correction.
Sufficient qubit fidelity to support efficient quantum error correction (QEC).
Achieving quantum error correction efficiently and accurately — even for a relatively small number of qubits.
Achieving enough physical qubits to achieve a sufficient capacity of logical qubits through quantum error correction.
Achieving sufficiently fine granularity of quantum phase and probability amplitude to enable quantum Fourier transform (QFT) and quantum phase estimation (QPE) for a reasonably large number of qubits sufficient to enable quantum computational chemistry.
Conceptualizing — and developing — a diverse and robust library of high-level algorithmic building blocks sufficient to enable complex and sophisticated quantum algorithms and applications.
Achieving The ENIAC Moment for even a single production-scale application.
Achieving The ENIAC Moment for multiple production-scale applications.
Conceptualizing — and developing — sufficiently high-level programming models to make quantum algorithm design practical for non-elite teams.
Conceptualizing — and developing — high-level quantum-native programming languages to support the high-level programming models using high-level algorithmic building blocks to make it easier for non-elite teams to design quantum algorithms and develop production-scale quantum applications.
Achieving The FORTRAN Moment for a single production-scale application.
Achieving The FORTRAN Moment for multiple production-scale applications.
Opening the floodgates for exploiting The FORTRAN Moment for widespread production-scale applications.

Each and every step in the progression could present its own quantum winter. Not necessarily, but it’s possible.

The only sure defense against such quantum winters is to sufficiently fund quantum computing research to enable enough redundancy and competition so that inevitable technical stumbling blocks and setbacks don’t bring all progres to a screeching halt as each roadblock is painstakingly analyzed and worked around, and sometimes even significant backtracking to find alternative routes. Better to pre-fund a plethora of alternative routes in advance so scientists and engineers always have choices available to them if promising approaches simply don’t work out, or don’t work out in a desired manner and on a desired schedule.

But occasional Quantum Springs and Quantum Summers

Even if we do indeed see one or more Quantum Winters, it’s just as likely that we will also see Quantum Springs and Quantum Summers as well, where despair turns around into hope and even euphoria.

In fact, we need to prepare in advance, not for the Quantum Winters, when things slow down, but for the Quantum Springs and Summers when we need to redouble our efforts to take advantage of the sun when it is shining. Or, as the old saying goes, Make hay when the sun shines, since the sunny days can pass by a lot faster than we might hope. So we need to have money, people, and resources in place well in advance of when the sun is about to start shining. In particular, a well-funded research program so that research ideas are ready to go and be turned over to product engineering when the sun starts shining.

Quantum Falls?

Uhhh… yeah, the downside of a Quantum Summer is a quantum fall setting us up for a Quantum Winter.

The only real defense for a Quantum Fall is to try to make the best of the downtime and available resources and talent to get ready for the next Quantum Spring and Summer.

Quantum Winters are a great time to double down on theory and basic fundamental research, which take a lot of time anyway. And a Quantum Fall is the best time to do the preparatory work for a Quantum Winter.

In short, plan ahead and minimize the chance that you will be caught unawares.

In truth, it will be a very long and very slow slog

In truth, quantum computing is actually advancing at a relatively slow pace — it’s the hype that is zooming ahead, breaking all speed records.

Sure, occasionally we will see dramatic breakthroughs, but usually we will see slow, steady incremental progress interspersed with modest to moderate-sized deserts or jungles that slow progress even further, even if on more rare occasions we see those amazing breakthroughs.

So, the basic reality is that we all need to be prepared for a very long and very slow haphazard slog of steady but uneven progress, even as we need to remain prepared for occasional bouts of euphoria… and despair.

No predicting the precise flow of progress, with advances and setbacks

Sure, everybody wants a schedule, a timeframe, and a detailed road map indicating when each milestone will be reached, but that just isn’t possible. Oh, sure, you can actually do that, but it will be obsolete before the ink is dry.

Some progress will be quite rapid, even while at other times progress will move at a snail’s pace, and with lots of occasional setbacks interspersed with occasional big advances.

Is a Quantum Winter likely in two to three years? No, but…

I do think that enough progress is being made on most fronts of quantum computing to achieve many, most — a critical mass, if not all of the advances needed over the next two years enumerated in the paper cited below and in a preceding section. If all of this transpires, then a so-called Quantum Winter is rather unlikely in two years.

Read my informal paper on these needed advances:

Why might a Quantum Winter transpire? Basically a Quantum Winter would or could commence due primarily to disenchantment and disillusionment due to failure of the technology to fulfill the many promises made for it — that quantum computing is not solving the types of problems that it was claimed to solve. So, as long as the vast bulk of the promises (a critical mass) are fulfilled, there should be no Quantum Winter.

But, fulfilling all of those promises is by no means a slam dunk.

But, a smattering of unfulfilled promises amid a sea of fulfilled promises would not be likely to be enough to lead to the level of disenchantment and disillusionment that would trigger or fuel a Quantum Winter.

So, the question is what the threshold balance is between unfulfilled promises and fulfilled promises which constitutes a critical mass sufficient to dip the sector into a Quantum Winter. That’s a great unknown — we’ll know it only as it happens in real-time since it relies on human psychology and competing and conflicting human motivations and behaviors.

If a Quantum Winter were to transpire, it would have some stages:

Critical technical metrics not achieved. Such as low qubit fidelity and connectivity, or still no 32 to 40-qubit algorithms.
Quantum Winter alert. Early warning signs. Beyond the technical metrics. Such as projects seeing their budgets and hiring frozen.
Quantum Winter warning. Serious warning signs. Failure of some high-profile projects.
Quantum Winter. Difficulty getting funding for many projects. Funding and staffing cuts. Slow pace of announced breakthroughs.
Severe Quantum Winter. Getting even worse. Onset of psychological depression. People leaving projects. People switching careers. Conferences being canceled.

Critical technical gating factors which could presage a Quantum Winter in two to three years

Some of the most critical technical gating factors which could send the sector spiraling down into a Quantum Winter in two to three years include:

Lack of near-perfect qubits.
Still only nearest-neighbor connectivity for transmon qubits. Transmon qubits haven’t adapted architecturally.
Lack of reasonably fine granularity of phase. Needed for quantum Fourier transform (QFT) and quantum phase estimation (QPE) to enable quantum advantage, such as for quantum computational chemistry in particular.
Less than 32 qubits on trapped-ion machines.
Dearth of 32 to 40-qubit algorithms.
Failure to achieve even minimal quantum advantage.
Rampant and restrictive IP (intellectual property) impeding progress. Intellectual property (especially patents) can either help or hinder adoption of quantum computing.

In short, I’m not predicting a Quantum Winter in two years and think it’s unlikely, but it’s not out of the question if a substantial fraction of the goals from this paper are not achieved — or at least giving the appearance that achievement is imminent.

Ultimately, a Quantum Winter is about psychology — whether people, including those controlling budgets and investment, can viscerally feel that quantum computing is already actually achieving goals or at least palpably close to achieving them.

Some technical walls could get hit

A lot of technical obstacles turn out to be mere speed bumps — minor inconveniences and quickly left behind and forgotten, but some turn out to be hard walls, at least in the near term it might seem that way. Some examples:

Qubit fidelity.
Qubit connectivity.
Coherence time.
Granularity of phase and probability amplitude.
Limits to quantum Fourier transform (QFT) and quantum phase estimation (QPE) precision.

I do expect that we will see advances in all of these areas over the next two to three years, but we could hit obstacles along the way that at least feel like hard walls, at least in the near term. Eventually we will get past any such obstacles, but the open question whether we will get past them before it is too late to avert a Quantum Winter two to three years from now.

Probably yes, but maybe not. That’s why we need to be cognizant of the potential for a Quantum Winter in the next two to three years.

Will trapped-ion qubits and neutral-atom qubits save the day?

Transmon qubits are having difficulty with low qubit fidelity and weak qubit connectivity. These may prove to be barriers to commercial success. But trapped-ion and neutral-atom qubits might be able to save the day since they have :

Generally higher qubit fidelity.
Full any to any qubit connectivity.

So, possibly they might be able to transcend the limitations of transmon qubits. One can always hope.

Still, they remain risky bets since their qubit fidelity is not that much higher.

So it remains to be seen whether either technology really can save the day.

Will new and innovative qubit technologies appear on the scene and save the day?

It’s definitely not outside the realm of possibility to anticipate that a variety of new and innovative qubit technologies might appear in research laboratories over the next two to three years which could change everything. But that’s not a slam dunk or even a remote certainty. Still, one can always hope, and the prospect is quite plausible.

Premature commercialization is probably the single biggest risk for stumbling into a Quantum Winter

I hate to have to say it, but quantum computing technology simply isn’t ready for prime time — it’s not ready for development and deployment of production-scale quantum algorithms and quantum applications. A premature effort to attempt to develop and deploy algorithms and applications with quantum computing technology in its current state is the single best recipe for driving ourselves into a deep Quantum Winter.

Instead, we should be focused on perfecting the technology so that it actually will be ready for commercialization in a few years. I call this pre-commercialization.

For more on pre-commercialization and the risks posed by premature commercialization, see my paper:

Might we simply sleepwalk into a Quantum Winter?

We may indeed simply sleepwalk into a Quantum Winter, not intentionally, but somewhat negligently, not being consciously aware and mindful of the growing gap between expectations (and hype) and reality.

Different audiences and sectors will experience a Quantum Winter differently or maybe even not at all

Not everybody will experience a Quantum Winter the same exact way. In fact, some may not experience it at all. Or, maybe, only some experience the full brunt of the Quantum Winter but most people don’t.

Different audiences have different interests and different exposures to features, capabilities, and risks.

Researchers are not likely to experience a Quantum Winter the same way as quantum application developers. And to some degree research may be able to stay out of the fray where all of the hype and exaggerated activity which leads to the Quantum Winter is occurring.

Different business sectors may experience a Quantum Winter differently as well, or not at all.

Some people who are at the leading or bleeding edge of the technology might experience a Quantum Winter much more acutely, while some at the fringes or trailing edges of the technology might be impacted to a much lesser degree — or even not at all.

Risk of Quantum Winter is primarily for those engaged in premature commercialization — those involved with research and pre-commercialization should be fine

Those involved in research, prototyping, and experimentation — what I call pre-commercialization — and running on minimal budgets — should feel relatively safe and secure from any effects of a Quantum Winter.

It is really only those who have drunk the Kool-Aid and are attempting to develop and deploy commercial, production-scale quantum applications — on large budgets — who are at great risk of succumbing to a deep Quantum Winter.

Quantum Ready — All dressed up and no place to go

Two to three years from now hundreds or even thousands of organizations with tens or even hundreds of thousands of technical staff will be officially Quantum Ready, but without the hardware, simulators, or rich enough programming models to be even remotely close to being ready to develop production-scale practical real-world quantum applications.

It’s one thing to prepare for the future, but it’s another thing to bet on when that purported and imagined future will transpire.

There is a significant risk or even likelihood that two to three years from now there won’t have been The ENIAC Moment yet for people to look to for inspiration and technical guidance.

We’re still in the full-on bliss of the honeymoon, but for how long?

So much of the technology of quantum computing is very new, fresh, and exciting, even intoxicating, but for how long will this honeymoon period last?

The Kool-Aid does taste so good. And there’s so much of it! Eventually it will spoil, sour, or just simply run out. What then?

But for now, you’ll have a very hard time convincing anyone to do anything other than to party on! It will end when it ends, and it doesn’t look like it will end real soon.

Saving grace: Nobody is calling quantum a mania or a bubble, yet

There is one significant saving grace: so far, nobody is calling quantum a mania or bubble, at least not yet. But let’s check back in a year.

What will assure that we can avoid this impending Quantum Winter?

There can be no absolute assurance of averting this impending Quantum Winter in two to three years, especially since human behavior and psychology are major factors, but at least from a technical perspective a Quantum Winter can be avoided two to three years from now if we can achieve…

The ENIAC Moment. A production-scale practical real-world quantum application can be demonstrated.
Minimal quantum advantage is achieved. At least 1,000X a classical solution.
Four nines of qubit fidelity. 99.99% reliability of two-qubit quantum logic gates and measurements.
Full any to any qubit connectivity.
At least one million gradations of phase and probability amplitude. If not a billion.
Quantum Fourier transform (QFT) and quantum phase estimation (QPE) for at least 20 if not 30 qubits.
Circuit depth of 2048 gates. Or is that still too low?
48 to 64 qubits with all of these qualities.

Additional requirements may be needed, but if this set is in place, we are well on our way to avoiding a Quantum Winter in two to three years.

Even minimal quantum advantage may remain out of reach in two to three years, but achieving it could avert a Quantum Winter

Dramatic quantum advantage is the holy grail of quantum computing — performance far beyond that of a classical computer, but it remains elusive and is presently out of reach and unlikely to be within reach within two to three years. Unlikely, but not totally out of the question.

Failure to achieve significant quantum advantage will remain a giant question mark hanging over quantum computing.

But if we can in fact achieve even minimal quantum advantage over the next two to three years, that would be a very good sign that could well prevent the onset of a Quantum Winter.

Dramatic quantum advantage (one quadrillion X a classical computer) is almost uncertainly out of reach of the next two to three years, but significant quantum advantage (1,000,000X) would be plenty good enough for that timeframe, and even minimal quantum advantage (1,000X) would be acceptable in this timeframe.

This is not to say that any of the technical goals for quantum computing are unimportant, but simply that most if not all of those technical goals must be achieved to achieve quantum advantage.

For more on dramatic quantum advantage, see my paper:

For more on significant quantum advantage and minimal quantum advantage, see my paper:

Single best way to avoid a Quantum Winter: Hold off on commercialization but double down on pre-commercialization

Premature commercialization may be the single largest factor which might cause the onset of a Quantum Winter, so the single best way to avoid the onset of a Quantum Winter is to hold off on premature commercialization while we focus all attention on much-needed research, prototyping, and experimentation, all of which are very needed as a predicate for eventual commercialization. I call this pre-commercialization, the work which must be completed before we even start thinking about actual commercialization.

It is not my goal to discourage interest or commitment to quantum computing, but to try to refocus people on pre-commercialization, focused on research, prototyping, and experimentation, rather than mindlessly forging ahead into premature commercialization which will only set us up for massive disappointment, disenchantment, disillusionment, despair, and depression — and a deep Quantum Winter.

For more on this model, see my paper:

Focus on research

A renewed focus on research will both enhance the quality of quantum computing technology, and take attention away from premature commercialization.

For more on needed research, see my paper:

Wildcards

There are any number of advances which could completely turn the tide and render a Quantum Winter much less likely, including:

New qubit technologies. Possibly with qualities not yet imagined.
New transmon architectures. Like some sort of dynamic bus for quantum state.
Dramatic improvements in trapped-ion qubits. And architectures.
Neutral atoms could address all of the concerns. Currently a great unknown, just great promises.
Distributed quantum computing and quantum networking. Computing with distributed quantum state could change everything.

What single advance within three years could turn the tide?

A dramatic improvement in fine-grained qubit phase and probability amplitude could have the largest impact on whether quantum computing slides into a deep Quantum Winter in two to three years.

Quantum Fourier transform (QFT) and quantum phase estimation (QPE) are the most powerful algorithmic tools available (so far!) in quantum computing. They also require higher qubit fidelity, greater qubit connectivity, and longer coherence time and greater circuit depth, but the critical factor for them is the need for fine-grained qubit phase and probability amplitude.

How fine-grained? A million gradations at a minimum — representing 20 qubits. A billion would be a better choice — representing 30 bits. And that’s just for the next two to three years. Even finer-grained phase and probability amplitude are needed further down the road.

With great support for quantum Fourier transform and quantum phase estimation we should definitely be able to avoid sliding into a deep Quantum Winter since they will enable amazing quantum computations, such as for quantum computational chemistry. Without them, a deep Quantum Winter seems a lot more likely since we would then be stuck with much more limited computational capabilities.

I would rank fine-grained qubit phase and probability amplitude top on the list of areas needing the most research attention. Sure, qubit fidelity, qubit connectivity, coherence time, and circuit depth also merit great research attention, but unless all of the pieces for quantum Fourier transform and quantum phase estimation are in place, quantum computing will be too limited to avoid a deep Quantum Winter in two to three years.

Configurable packaged quantum solutions are the greatest opportunity for widespread adoption of quantum computing

Although this might not be strictly required to avoid a Quantum Winter in two to three years, an interest and focus on configurable packaged quantum solutions might go a long way towards building positive excitement and enthusiasm, and focus on the critical technical deficits which need the most research attention, going a long way towards eliminating many of the technical deficits that might otherwise send quantum computing down into a Quantum Winter.

For more on configurable packaged quantum solutions, see my paper:

Roadmaps would help avoid disappointment — if they are informative and reasonably accurate

Not all hardware (or software) vendors publish roadmaps and milestones for their products. Even when they do, the level of detail and accuracy can vary greatly.

An informative and accurate product roadmap can work wonders to set reasonable expectations, and hence work wonders to avoid sliding into a deep Quantum Winter.

On the flip side, lack of roadmaps, minimal detail, and inaccuracy are great ways to slide into a deep Quantum Winter.

I won’t attempt to critique all vendors here, but I did write up a critique for IBM’s hardware roadmap (and a little about their quantum ecosystem roadmap as well):

The bottom line there is that IBM does provide some detail and milestones, but not enough detail on some of the most critical hardware shortcomings, so that it doesn’t lead me to believe that they will definitively be able to avoid sliding into a deep Quantum Winter in two to three years.

My personal disappointments

This is a short list of the my own top priorities for advances which are holding quantum computing back and could lead to a Quantum Winter if not reasonably addressed over the next two to three years:

Qubit fidelity. Not clearly stating qubit fidelity as top priority.
Near-perfect qubits. Not focusing on the need for near-perfect qubits.
Qubit connectivity. Not focusing on weak qubit connectivity of transmon qubits.
Quantum Fourier transform (QFT) and quantum phase estimation (QPE). Not focusing on quantum Fourier transform (QFT) and quantum phase estimation (QPE) and their need for fine-grained phase and probability amplitude.
Disappointment with IBM 127-qubit Eagle. Plenty of qubits, but only a quantum volume of 64. Mediocre qubit fidelity. Weak qubit connectivity. Hardly better than the 27-qubit Falcon.
Weak documentation and specifications. General lack of detail and precision.
Lack of a high-level programming model.
Lack of a rich collection of high-level algorithmic building blocks.
Not pushing and enhancing simulators strongly enough. Need for performance, capacity, configuration, and debugging, testing, and analysis tools.
General lack of scalable quantum algorithms. Doesn’t seem to be a priority.
Premature commercialization. Far too much “drinking the Kool-Aid” and engaging in premature commercialization.
Research. Not enough focus on research.
Pre-commercialization. Need to shift away from commercialization, back to pre-commercialization, with emphasis on research, prototyping, and experimentation.
Much research remains to be done. Research is beginning to suffer as too many people focus on commercialization.
No quantum applications yet. A great promise for the future, but not realized in any way yet.
Configurable packaged quantum solutions. Best focus for implementing general-purpose quantum applications. Would focus attention on areas needing research, address the major technical deficits, and help build positive excitement and enthusiasm which could avoid a Quantum Winter.

No quantum applications yet — a great unrealized promise

We have plenty of fragments of quantum algorithms floating around, published in plenty of academic papers, but essentially no realization of the great promise of a future filled with all manner of quantum applications.

It will be rather difficult to avoid sliding into a deep Quantum Winter if we don’t have a wide range of quantum applications in two to three years.

It’s certainly not out of the question to see lots of quantum applications developed over the next two to three years, but that’s a heavy lift and I don’t see a clear and obvious path to that outcome in that timeframe. There are too many technical deficits to overcome, and too little expertise for designing practical and production-scale quantum algorithms and quantum applications.

For a general sense of the unrealized promise and the types of quantum applications that have been promised, see my paper:

General lack of scalable quantum algorithms and applications — they don’t seem to be a priority

Quantum algorithms and quantum applications need to be scalable. For a variety of reasons. But, commonly, they’re not. It doesn’t seem to be a priority for most people and most organizations.

We need to be able to rapidly — even instantly — be able to exploit new quantum computers with more qubits without the need to redesign the quantum algorithms or quantum applications. If we can’t do that, it will be difficult and expensive to break out from small toy-like quantum computers to the real, production-scale quantum computers needed to avoid slipping into a Quantum Winter in two to three years.

For more on scalability of quantum algorithms and applications, see my paper:

Will Shor’s factoring algorithm be able to factor even 8 to 12-bit semiprime integers?

Although no one expects that Shor’s factoring algorithm will be able to factor any large semiprime integers in two to three years (say 1024 or 512 or 256 or even 64 bit integers), it would still be rather disheartening if we can’t factor even 8 to 12-bit semiprime integers two to three years from now. Factoring semiprime integers won’t be a high priority during this period, but it will be yet another factor that will work to undermine confidence in the near-term power of quantum computing. It probably wouldn’t be enough to trigger a Quantum Winter by itself, but it could be a significant factor if the quantum hype continues to focus excessive attention on the power and promise of Shor’s algorithm.

Shor’s factoring algorithm uses quantum Fourier transform (QFT) and quantum phase estimation (QPE), so it actually is a decent surrogate for evaluating all of the major technical factors that will determine whether we spiral down into a deep Quantum Winter in two to three years, namely:

Need for higher qubit fidelity.
Need for greater qubit connectivity.
Need for finer granularity of phase and probability amplitude.
Need for quantum Fourier transform (QFT) and quantum phase estimation (QPE).
Need for greater circuit depth. Need for greater coherence time and/or faster gate execution time.

All of those factors will determine how workable Shor’s factoring algorithm will be in two to three years. And, as a happy coincidence, these same factors will determine the viability and capability of quantum computing in general in two to three years, hence guiding whether we lean towards or away from a Quantum Winter in two to three years.

So, we won’t need Shor’s algorithm per se, but inability to use Shor’s algorithm to factor even 8 to 12-bit semiprime integers in two to three years will indirectly bode poorly for quantum computing overall.

On the flip side, if Shor’s algorithm can be implemented to factor 24 or even 16-bit semiprime integers in two to three years, quantum computing will be seen as being in decent shape and on track towards enabling production-scale practical real-world applications in the relatively near future (three to five years), and hence facilitating avoidance of a Quantum Winter in two to three years.

In short, implementation of Shor’s factoring algorithm is not a mandatory metric, but a very telling metric nonetheless.

When might a Quantum Winter start?

Trying to precisely estimate the exact date for the onset of a Quantum Winter is a fool’s errand, but we can assess some likelihoods:

Six months. Nope.
One year. Very unlikely.
18 months. Unlikely. But it is possible.
Two years. Possible.
Two and a half years. Very possible.
Three years. Very possible.
Four years. Possible.
Five years or longer. Possible.
Seven years or longer. Possibly a second Quantum Winter.

How long might a Quantum Winter last?

A Quantum Winter will last until some trigger event that ends it (see the next section). The duration of a Quantum Winter could range from:

A few months.
Six months.
Less than a year.
One year. Possibly. Likely longer.
18 months. Possibly.
Two years. Maybe more likely.
Two and a half years. Maybe more likely.
Three years. Maybe more likely.
Four to five years. Less likely. Shouldn’t be that long.
Five to seven years.
Seven to ten years.
Even longer, maybe.

What prospect are we presenting facing? Possibly a Quantum WInter of two to three years. It wouldn’t have to be that long technically, but it’s back to the human psychology factor — once depression sinks in, it just takes a protracted period of time to restore confidence.

It takes a protracted period of time to restore confidence

Just to emphasize this point more strongly, once confidence has been undermined and broken, which can happen quickly, it can take a lot longer to restore confidence — even once the technical factors have actually been addressed.

There will likely be a lot of psychological resistance. And a lot of bad memories to be processed and gradually pushed aside.

Even once a few key technical successes are achieved, such as The ENIAC Moment, it could take some time for many people to accept that the accomplishments by a super-elite team are also possible by less-elite technical teams.

Some successes with configurable packaged quantum solutions might be needed to get many people over this psychological hump.

What ends a Quantum Winter?

A Quantum Winter continues indefinitely until one or more trigger events, which can include:

A basic advance. Something simple but essential.
A technological breakthrough innovation. Something really dramatic.
Availability of funding.
Appearance of a problem which requires the technology for its solution.
Entry of new persons with fresh new motivation and fresh ideas.

Innovation to end a Quantum Winter

The mere existence of an innovation won’t necessarily immediately end a Quantum Winter. It may have to move through stages, a process:

The innovation is conceptualized. The idea.
The innovation is realized. Implementation.
Prototyping with the innovation.
Experimentation with the innovation.
Decision to adopt the innovation.
Process of adopting the innovation.
Promotion of the innovation.
Acceptance of the innovation.
Practical adoption of the innovation.
Gradual reacceleration of investment, budgets, projects, and products based on the innovation.

An example innovation is the transistor, which took a decade until it was practical enough for commercial production of computers.

Why haven’t we seen a Quantum Winter yet even though past progress over the past 25 years was frequently quite slow?

It’s not the raw rate of actual progress which triggers a Quantum Winter, but excessive hype and irrational expectations relative to reality.

Five to 25 years ago we didn’t have vast hordes of Kool-Aid-drinking fans with outrageously-high short-term expectations and a media pumping out hype and setting irrational expectations.

We’ve had quantum slumps or slowdowns in the past, but no large commercial budgets or venture investments to be impacted. Research money is much more patient and focused on the long-term with very slow, incremental progress.

People and money will be on an increasingly short attention and patience span as we move through the next two years.

Right now, people are still reasonably patient. They’re more in awe of the technology and its exotic promises than expecting short-term results.

But in a year to 18 months there will be clamoring for short-term results that will become a terrible cacophony which won’t be easily placated. The mob will (okay, might — I’m speculating) want blood and settle for nothing less.

So we have had no Quantum Winter so far since there is no excessive level of irrational exuberance to deal with — so far, but wait a year to 18 months.

So far, we’ve been focused on smaller algorithms with few qubits, with no quantum Fourier transform (QFT) or quantum phase estimation (QPE) or deep circuits, so we haven’t been confronting the full depth of the problems which will push us into a Quantum winter in two to three years as we do try to move deeper into more complex and more sophisticated quantum algorithms and quantum applications.

We’ll likely not spiral down into a Quantum Winter until we attempt a serious move into 40-qubit algorithms, which we haven’t done yet.

My original proposal for this topic

For reference, here is the original proposal I had for this topic. It may have some value for some people wanting a more concise summary of this paper.

Risk Is Rising for a Quantum Winter for Quantum Computing in Two to Three Years. Be prepared for a possible Quantum Winter in 2–3 years as promised and expected technology just isn’t ready yet or as capable and easy to use as expected. Many organizations will have invested significant resources with an expectation of getting a dramatic return on their investment by then. Investors, executives and managers may begin losing patience, especially as planners are unable to project how many more years and additional resources before the technology really is ready for production-scale deployment.

Summary and conclusions

Quantum Winter is unlikely.
Quantum Winter is not imminent in the coming months or year.
A Quantum Winter is unlikely in two to three years based on my personal expectations for progress, but… I’m an optimist, but… I’m also a pragmatic realist and the risks are rising.
But quantum winter is a very real risk two to three years out.
Risk seems to be increasing. Rising concerns. As hype and expectations rise faster than reality.
The tsunami is building slowly — but surely. Fueled by the raw energy of the massive hype and exaggerated and unrealistic expectations. Or large expectations which may be met, but might not be fully met. Maybe progress will be deep enough in two years that the energy of the tsunami will be distributed widely and not in some great wave that triggers a deep Quantum Winter which wipes out the field for many months or even years.
Patience will be the key factor determining the onset of a quantum winter. Those with great patience will not be causing a Quantum Winter. Patient capital will prevail.
A Quantum Winter is largely psychological in nature. Sure, it’s based on issues with the technology, but it is the impact on human psychology that permits the technical issues to drive people into a Quantum Winter.
Sentiment matters — positive sentiment is needed.
Next year should be fairly clear sailing.
Maybe some hint of rising risk in 15 to 18 months or so.
If we do see a Quantum Winter it will be after we spend two years to presumably get to being on track for production-scale quantum applications, and then one year for disappointment and disenchantment to unfold into deep depression if it turns out that we are not on track after two years from now.
Palpable sense that all is not okay in two years.
Either progress accelerates in second year — and third year — or feeling of impending quantum winter takes root and takes off.
If a quantum winter does take root, figure two to five years before subsequent Quantum Spring.
Quantum computing is still at the stage of being a mere laboratory curiosity, not even close to being ready for development and deployment of production-scale practical real-world quantum applications.
Still more appropriate for the lunatic fringe rather than mainstream application developers.
Single biggest risk for triggering a Quantum Winter is premature commercialization.
People are starting to drink way too much of the marketing Kool-Aid. The elixir of impending commercialization is beginning to take root. And people are reluctant to commit to the hard work and slow slog of pre-commercialization.
Single best way to cause a Quantum Winter is to pursue premature commercialization before the technology actually is ready to develop and deploy production-scale practical real-world quantum applications.
Single best way to avoid a Quantum Winter is to avoid premature commercialization and double down on per-commercialization instead, particularly research, as well as prototyping and experimentation.
Everything we can do to get people much more focused on research, prototyping, and experimentation rather than commercialization will help to avoid falling into a deep Quantum Winter caused by premature commercialization.
Many or most advanced technologies inevitably experience seasonal cycles of some sort, with one or even more technological winters, so it may be unrealistic to expect that quantum computing won’t have its own share of seasonal cycles.

For more of my writing: List of My Papers on Quantum Computing.