Hide table of contents

Cross-posted from the High Impact Engineers Resource Portal. You can view the most up-to-date version on the Portal.


Risks posed by nuclear war remain some of the most serious facing humanity today. Despite lower public and philanthropic interest in recent years, the risk of nuclear war persists. The outcome of a nuclear exchange of any kind would be disastrous, even if the exact consequences are difficult to model. A large nuclear exchange poses a risk of collapse of civilisation or perhaps even human extinction, via nuclear winter and vulnerability to subsequent threats.

Thankfully there are strategies to mitigate these risks. The pursuit of these strategies has led to wins such as a dramatic reduction in nuclear stockpiles since the height of the Cold War. Engineers can leverage their technical expertise to promote these strategies through:

  • Research, e.g. through analysis of satellite data
  • Technical work, e.g. through improving verification technologies for arms control
  • Policy, e.g. through programs that train STEM professionals on global security to provide advice for policymakers

There are several areas of work that engineers should avoid when trying to reduce the risks posed by nuclear weapons. These include improving missile defence systems, advancing nuclear arms capabilities and doing work that poses significant information hazards (e.g. publishing technical information on weapons).


The content of this article is largely based on research by 80,000 Hours, the Nuclear Threat Initiative, and the Center for Arms Control and Non-Proliferation. Although funders, experts, and decision-makers face deep uncertainty about the effectiveness of different interventions, we feel reasonably confident that the recommendations in this article are robust.

Cause area overview

Risks from nuclear war

Likelihood of nuclear war

Though tensions between large nuclear powers and global weapons stockpiles have decreased since the height of the Cold War, risks of nuclear war persist. Nuclear powers have avoided nuclear use after World War II by adapting strategies of deterrence, where a nuclear power’s arsenal is used to deter other powers from being the first to use their own. Maintaining stable deterrence has become crucial and is dependant on factors such as each actor credibly making use of their weapons, having a second strike capability, and that each actor is vulnerable to a strike for another.

Today there are several destabilising factors that are a cause for concern:

  • Nuclear escalation from ongoing conflicts and potential future conflicts, such as in Ukraine, Taiwan, or Kashmir.
  • China’s expansion of its nuclear arsenal and the unprecedented issues of “three-way deterrence” between Russia, China, and the United States
  • Technological advancements including cyber warfare, artificial intelligence, missile defence systems and advanced delivery systems.
  • National warning systems that are vulnerable to malfunction or human error, causing nuclear retaliation to a non-existent oncoming strike.
  • Internal political instability or poor political leadership leading to first nuclear use.

Due to the complex nature of the problem, the likelihood of nuclear use is extremely difficult to estimate. Estimates of the probability of a nuclear conflict between the U.S. and Russia (the most destructive of any two nation conflict)  range between 0.01% and 2.21% per year. Assuming the level of risk remains constant, this would imply the likelihood of nuclear war between these powers to be 10–85% in the next 100 years.

Outcomes of nuclear war

Any use of nuclear weapons would be devastating to the receiving nation and would be highly destabilising to international relations. Today there are “low yield”, “tactical” or “battlefield” weapons that leaders have considered deploying in conventional conflicts. Many of these weapons are much more powerful than those used in Hiroshima and Nagasaki. The use of these weapons has left an indelible mark on Japan and the global community, demonstrating that even an extremely limited use of nuclear weapons can have disastrous effects.

Large-scale nuclear war

Nuclear arsenals today are much larger and also include much higher-yield “strategic” weapons capable of destroying dense cities and fortified missile silos.  Casualties are difficult to estimate, but an all-out nuclear exchange between the two largest nuclear powers today would directly kill tens of millions of people, depending on the sites targeted. These deaths would result from both the initial blast and deaths from nuclear fallout in the following weeks.

In 2019, Rethink Priorities conducted in-depth research into the risks from nuclear weapons. They predict a nuclear exchange between Russia and the US would have the highest potential of harm due to three factors: 1) the size of the countries’ nuclear arsenals; 2) the size of the countries’ populations; and 3) the probability of the given nuclear exchange scenario. Nuclear exchanges between India and Pakistan, and China and either the US, India, or Russia would likely be the next most devastating.

According to this model created by Rethink Priorities, the number of people that would likely be killed directly by nuclear detonations during a US-Russia nuclear exchange is about 51 million. This would absolutely be devastating, but it’s important to keep scope sensitivity in mind when trying to understand how important the threat of nuclear war is in comparison to other pressing problems. (To put that figure into context, more than 50 million people died from the Spanish flu pandemic of 1918–1919 and civilisation has continued). Current prevailing work on preventing nuclear war overlooks scope sensitivity, and focusing on the most devastating risks is one key way that people new to the field of preventing nuclear war can add value.

However, an all-out thermonuclear war between the US and Russia be a civilisation-ending events if the high estimates of probability and severity of the climate effects of nuclear winter are true.

Nuclear winter

The aftermath of nuclear war also poses potentially serious risks. Huge fires from a large nuclear exchange could send particulate matter into the atmosphere, which would block incoming sunlight. This would lower temperatures globally for up to a decade and reduce global rainfall by about 15-30%. Reduced photosynthesis and thermal energy could lead to ecosystem disruption, agricultural collapse and mass starvation.

Estimates of the likelihood and extent of nuclear winter effects are disputed. Some contemporary literature asserts that more modest climatic effects would be felt, which would not lead to mass death. It must be kept in mind that the models and empirical data sets in this field are limited and uncertain. While we should perhaps no longer think that a large exchange would most likely lead to extreme nuclear winter, its possibility should not be entirely discounted.

All-out nuclear wars are disproportionately worse than highly limited nuclear exchanges due to climate dynamics, threshold effects, and the possibility of societal collapse. This Founders Pledge report tentatively concludes that “not all wars would lead to nuclear winter, but larger wars are more likely to lead to severe climate change, and threshold effects amplify the superlinear cost of this.”

Even in the absence of a nuclear winter, a world after a large-scale nuclear war may be extremely vulnerable to subsequent extinction risks. This includes pandemics, biological warfare (see our page on Biorisk and Biosecurity), artificial intelligence, and other risks that might wipe out humanity or reduce us to a number too small to effectively rebuild.

Mitigation of Risks

There are various strategies for mitigation of the risks above. Many of these are primarily framed as policy interventions, but pursuing these strategies requires people with technical expertise. Due to the present situation of deterrence, many of these interventions aim to improve the outcome of a nuclear exchange without having a destabilising effect. Some have proven effective and have led to wins such as a dramatic reduction in nuclear stockpiles since the height of the Cold War.

Figure 1: Our world in data, Estimated nuclear warhead stockpile 1945 to 2022, Roser et al. (2022)

Any action that promotes cooperation between nuclear powers, or improves international relations in general, is also likely to reduce risks from nuclear war.


Disarmament is the process of reducing the quantity and/or capabilities of military weapons and/or military forces. In discussions of nuclear weapons, disarmament consists of reducing, limiting, or abolishing nuclear weapons, generally through international treaties and agreements.

Disarmament efforts have been extremely successful in reducing the overall risks of nuclear war, reducing current arsenals to a fifth of what they were a half-century ago. As successful disarmament efforts reduce the severity of outcomes in an all-out exchange, it is the best strategy for reducing the severity of the worst possible nuclear outcomes.

Some disarmament efforts include the aim of reducing weapon yields. However, low-yield weapons can be seen as a lesser escalation and have fewer controls on their use, potentially making nuclear use more likely. These weapons would still be similar in yield to those used in Hiroshima and Nagasaki and may prompt further escalation.

Efforts to reduce weapon yield therefore remain somewhat controversial, despite reducing overall nuclear capabilities. We believe they are likely net positive as low-yield weapons would cause less destruction in a large nuclear exchange and many of the mechanisms of nuclear escalation remain uncertain.


Nuclear non-proliferation efforts focus on preventing the spread of nuclear weapons and their delivery systems. Non-proliferation efforts have had mixed success, with the number of nuclear armed countries rising significantly since the inception of nuclear weapons.

Figure 2: Our world in data, Number of countries that consider, pursue, or possess nuclear weapons, Roser et al. (2022)

The likelihood of a nuclear exchange is thought to be greatly increased when more actors with conflicting priorities possess nuclear capabilities. Therefore successful non-proliferation efforts would reduce the chances of a nuclear exchange occurring.

The vast majority of nuclear firepower is held by the United States and Russia. As non-proliferation is not directly aimed at reducing the likelihood or severity of nuclear exchanges between these powers, this intervention may not be effective in reducing the risks from the largest possible nuclear exchanges. However, reducing the destabilising effects of creating additional nuclear powers may be relevant in reducing the chances of a US-Russia exchange.

Banning nuclear testing

Test ban treaties such as the Comprehensive Nuclear-Test-Ban Treaty aim to eliminate the full scale testing of nuclear weapons. Though some nuclear powers have not formally pledged to ban testing, international norms against testing has led to very few tests being carried out since the end of the Cold War.

Figure 3: Wikimedia, Participation in the Comprehensive Nuclear-Test-Ban Treaty (2022)

One notable exception to this is tests carried out by North Korea (DPRK), which is the only nation to have continued testing nuclear arms into the 21st century. The most recent of these tests was in 2017, however over the past few years they have also carried out many tests of their new missile technologies in close proximity to neighbouring countries.

Figure 4: Our World in Data, Number of nuclear weapons tests, 1945 to 2019, Roser et al. (2022)

Nuclear tests and tests of missile technologies are often a form of posturing and lead to rises in tension between nuclear powers. It also bears mentioning that nuclear tests are extremely harmful to local communities, sometimes causing generations of genetic disorders due to radiation exposure. Banning these tests would lead to a more stable political landscape.

Nuclear test bans are less popular among newer nuclear powers or nations pursuing nuclear arms. As established nuclear powers have data from past tests and usually have greater resources for other research such as supercomputer simulation, this puts them at a distinct advantage when expanding or modernising stockpiles.

No first use

Nations with a No First Use policy pledge to refrain from the use of nuclear weapons, except for as a second strike in retaliation. No First Use policies create more stable international relations by eliminating the ambiguity about a nation’s intentions with its nuclear arsenal.

A nation may decide to launch a first strike to maintain an element of surprise. A first strike will generally aim to destroy an adversary's military infrastructure or nuclear forces, reducing their ability to retaliate. This creates a “use or lose” situation where much of the nuclear arsenal would be lost if not launched upon detection of an oncoming strike. Eliminating the option of first use is unattractive to military tacticians as it puts their nations at a distinct disadvantage.

China and India are currently the only two nuclear powers to formally maintain a No First Use policy, adopting pledges in 1964 and 1998 respectively. Both NATO and a number of its member states have repeatedly rejected calls for adopting a No First Use policy, saying this would weaken their deterrence. Russia, Pakistan, North Korea (DPRK), and Israel have also not formally adopted a No First Use policy, although some have stated they would only deploy nuclear weapons in response to a large-scale conventional attack.

This failure of many nuclear armed nations to formally adopt a No First Use policy incentivises other nations to increase their deterrence by stockpiling or pursuing nuclear weapons programs. This impedes both disarmament and non-proliferation efforts.

No launch on warning

Launch on warning is a military strategy that allows the use of a retaliatory nuclear weapons strike against an opponent as soon as satellites and other warning sensors detect an incoming enemy missile.

As most intercontinental nuclear strike strategies aim to remove the adversaries’ capability to launch nuclear weapons, a strike of retaliation may need to be launched before the initial strike would land. This decision would have to be made in the order of minutes. Adopting this policy strengthens the deterrence of a nuclear power by guaranteeing damage to an aggressor.

This strategy is extremely vulnerable to false alarms of imminent strike, of which there have been notable examples over the relatively short history of nuclear weapons. Taking a policy of not launching on warning would avert the chances of a nuclear exchange prompted by radar malfunction or human error in data interpretation.

These policies would be difficult to enact as appearing to launch on warning is important in the deterrence of initial nuclear strikes. No countries have formally adopted a no launch on warning policy, although only NATO countries and Russia have strategic weapons deployed which would be capable of automated launch on warning.


De-alerting is the introduction of reversible changes to nuclear weapons systems, in order to lengthen the time required to launch these weapons. As many weapons systems remain on “high alert” in case of an oncoming strike, de-alerting has been proposed as a means to reduce the accidental or deliberate use of nuclear weapons.

De-alerting offers the opportunity for a lower buy-in form of disarmament. As the changes to the weapons systems are reversible, nations may be more likely to put them in place than to destroy existing weapons. This would reduce the chances of nuclear use while allowing nations to maintain their nuclear arsenals.

Reliable implementations of de-alerting would require intensive verification procedures. It is also very unpopular among military tacticians as it reduces a nation’s deterrence threat.

It is also worth noting that de-alerting alone would not reduce total weapon stockpiles, only launch cadences. This may reduce the total number of weapons used in the largest exchanges as fewer launch facilities would be used before being struck, depending on the nature of the exchange. Policies of disarmament are preferable when possible due to their reduction of stockpiles and more straightforward ongoing verification procedures.

No underground silos/submarines

Nuclear silos are hardened underground facilities for housing and launching ballistic missiles. These are often solid fuel intercontinental ballistic missiles that are on high alert, such as the U.S.’s appropriately named Minuteman missile. Having these missiles housed in silos means that an enemy must break each of these silos in a first strike if trying to limit the effects of a retaliatory strike. Each group of silos often takes two or more of their own ICBMs to break. To this end, nuclear powers often build many silos (which aren’t too expensive) and fill them with a few ICBMs (which are extremely expensive) to maximise the chances of having weapons left in a second strike.

Nuclear armed submarines (distinct from nuclear powered submarines) are functionally impossible to detect even when operating within enemy waters, making them extremely strategically advantageous. Their nuclear ballistic missiles can be deployed even in cases where a first strike has wiped out a nation’s entire land-based forces, making them an extremely effective deterrence force.

Figure 5: The United Kingdom’s Vanguard class submarine equipped with the Trident nuclear missile, UK Royal Navy Press

Eliminating or reducing the number of silos would significantly reduce the impetus for stockpile build-up in other nations, as fewer weapons would need to be used to eliminate an adversary's second strike potential. Building many silos also has a destabilising effect, as it can be interpreted as a signal of a nation’s intention to increase its nuclear arsenal significantly.

Nuclear armed submarines effectively present an immovable second strike threat, introducing a heightened level of uncertainty and instability in international relations. Banning their use would foster greater transparency and predictability, improving trust and reducing tensions among nuclear-armed states. Submarines are also more likely to experience communication issues or to be cut off from communication with command. These communication issues may lead to misuse of their arms and have led to near misses in the past.

The same trade-off between lowering nuclear risks and diminishing a nation’s deterrence can be seen here. It is extremely unlikely any nuclear power would elect to give up their silos or nuclear armed submarines in the foreseeable future because of the huge strategic advantages they offer.

Post-nuclear use, or "right-of-boom" interventions

The most deaths from an all-out nuclear war will likely come from climate effects from nuclear winter. As a result, there may be an argument for researching nuclear postures and capabilities that retain potential deterrent effects, while minimising the probability of nuclear winter. E.g., can yields, burst heights, targeting policies, etc. be adjusted in ways to minimise soot production while retaining deterrence?

Founders Pledge deep dives into more right-of-boom interventions in their report.

Civilisational resilience

Increasing the resilience of population centres, food supply, and infrastructure would reduce the overall death toll of a nuclear exchange. This would reduce the negative outcomes of a nuclear exchange without reducing a nation’s deterrence threat or destabilising international relations.

Further information on the topic can be found on our cause area page: Civilisation Resilience.

How can engineers do impactful work on risks from nuclear war?

Much of the impactful work in the field is focused on implementing the risk mitigation strategies listed above. Though the skills required for these roles are often in policy and security analysis, there are some bottlenecks where engineers are needed.

What are the bottlenecks?

Research, analysis, and educational work

Working as a researcher/analyst in non-profit or academic organisations

Academic institutions and non-profits such as think tanks play a vital role in shaping the information landscape and informing policy. Many also contribute to open source intelligence projects, which inform the public of pressing issues.

Engineers are well suited to work such as:

  • Analysis of satellite data and other images
    • Satellite data has become invaluable in the modern age. Analysts help governments and the open-source community calculate how many nuclear weapons a nation is producing, or if a country is abiding by an agreement not to produce fissile materials.
    • For example, in recent years commercial satellite imagery has been used to identify the rapid growth in the number of Chinese missile silos, discovered that North Korea has been producing additional fissile material and mapped the layout of new Russian nuclear facilities in the Arctic.
    • Other images of nuclear weapons and their supporting systems include those from military parades, propaganda material and news coverage. These can all be used to better characterise weapons systems and infer their capabilities.
    • Engineers with experience in data science, aerospace, manufacturing and nuclear power would be especially well-suited to this work.
Figure 6: Infrared satellite image of heat emitted from a DPRK (North Korean) facility, likely indicating further enrichment of fissile material for upcoming warheads, Center for Strategic and International Studies
  • Analysis of international trade.
    • Analysis of international trade has been a key tool for keeping the global community informed about nuclear proliferation and stockpile build-up. This includes electronics used for guidance, military vehicles that could be used to support mobile launches, industrial equipment used for processing fissile material, and countless other goods that could betray a nation’s nuclear intentions.
    • Technical expertise in many different facets of manufacturing can be pivotal in interpreting data on national imports.
    • A notable blunder in this analysis is the famous aluminium tubes imported by Iraq in 2001. It was theorised by the US that these would be used for building a uranium enrichment centrifuge. This was heavily leaned on as a justification for the subsequent invasion of Iraq. The tubes were actually destined to be turned into conventional artillery rockets but also could have been used for many other purposes. Better quality analysis could have led to a better outcome in this case; however, there are political factors at play that complicate this example.
    • Engineers with experience in manufacturing engineering, aerospace, and nuclear power would be especially well-suited to this work.

Non-profit organisations that hire for such work include:

Research work within the Effective Altruism community

The Effective Altruism community is a large network of largely non-profit organisations, of which High Impact Engineers is a part. There is a large intellectual and financial interest in global catastrophic risks, including those posed by nuclear weapons.

Much of the research on nuclear war within effective altruism is conducted by generalist analysts. This allows the research to be broad in scope but can be lacking when technical expertise is required. Researchers within Effective Altruism have identified technical research topics and larger research themes that would require technical experts to fill knowledge gaps. Engineers with a wide variety of backgrounds could contribute here, as technical experts are sparse within Effective Altruism.

Organisations within Effective Altruism that may offer opportunities for work in nuclear safety include Rethink Priorities, 80,000 Hours, the Centre for the Study of Existential Risk, the Future of Life Institute and the Swiss Existential Risk Initiative. It is also somewhat common in Effective Altruism for independent researchers to seek funding for smaller projects. Funding sources such as EA funds, Open Philanthropy, Lightspeed grants, Founders Pledge, and Longview Philanthropy may be interested in work on nuclear war.

Educating policymakers and the public about technical matters related to nuclear war

It is difficult to conceptualise something so powerful and complex as nuclear arms systems. For nations to take action to minimise the chances of nuclear use, policymakers and the public must be well-informed on relevant technical topics.

People with general technical expertise often have an advantage when consuming technical information and rapidly synthesising this into a format that would be digestible to someone with no background in the field. Technical expertise is often respected by policymakers who will heed expert judgment when enacting policy. This may be especially important today, with digital media allowing the rapid spread of misinformation.

Engineers with experience in research, journalism, policy or education (either secondary/high school or college/university level) would be especially well-suited to this work.

Technical work

Improving verification for arms control

Verifying compliance with arms control treaties requires people with technical expertise in areas such as fissile material, manufacturing technology, and aerospace engineering. This is historically achieved through verification envoys visiting other nations’ nuclear facilities to verify weapons numbers, amounts of fissile material, and facility security. Increasingly new technologies are being used to verify treaty compliance.

Verification of compliance with arms control treaties has recently become extremely important. Russia has suspended the inspection of its nuclear arms, previously agreed under the New START treaty. China is building its nuclear forces and is resistant to verification and inspection. Not having had the arms control history of the US-USSR, they are not as comfortable with letting inspectors roam their facilities.

New technical solutions may need to be found that give access to just enough information to confirm agreed terms, without incidentally gathering any additional information. Engineers with experience in aerospace, manufacturing, nuclear physics, and data analysis would be especially well-suited to this work.

Improving technologies to prevent theft of fissile material and nuclear weapons technologies

Theft of fissile material by nations or non-state actors would present enormous issues for non-proliferation efforts. It can advance national nuclear weapons programs, lead to nuclear terrorism, and contribute to political instability.

Radioactive sources including cobalt, caesium, and iridium isotopes are present worldwide. Many of these have civilian uses in medicine, energy, and research. Only 20 states have at least one kilogram of fissile materials like highly enriched uranium and plutonium which could be used for weapons. This figure is down from 40 states as recently as 2005.

It should be noted that while the nuclear arsenal of newly nuclear armed nations or terrorist groups could be enough to create significant human tragedy, they wouldn’t be sufficient to cause nuclear winters or threaten civilisation collapse. It’s possible that nuclear first use by these actors could precipitate further nuclear use in some scenarios, however.

There are also sensitive technologies that would present non-proliferation issues if stolen. This includes rocketry technology, uranium enrichment technology, and submarine technology. This may be theft of the technology itself, plans and technical drawings, or even just technical information communicated by a member of staff.

Additional technical solutions are needed worldwide to protect these materials and technologies. Computer systems may be vulnerable to cyberattack, facilities may be vulnerable to physical breach and additional controls could be placed on material and information.

One famous historical example of a breach of nuclear facilities is the commuter program Stuxnet’s destruction of many of Iran’s enrichment centrifuges. Stuxnet was a cyberweapon likely built jointly by the United States and Israel in a collaborative effort known as Operation Olympic Games. Although no theft of fissile material occurred, this example underscores the vulnerability of these facilities.

Engineers with experience in cybersecurity, electronics, and defence would be especially well suited to this work.

Improving the robustness of nuclear command, control, and communications systems

In the US, the nuclear command, control, and communications (NC3) system refers to the chain of authority and associated technologies that control nuclear use. Other nations have similar systems.

The proper functioning of these systems is vital to preventing the misuse or accidental use of nuclear arms. A new set of technological challenges will be faced in the coming years by cyber-attacks on NC3 systems, laser attacks on satellites in space, and artificial intelligence applied to autonomous weapons systems, such as submarine drones.

Engineers with experience in cybersecurity, communications systems, electronics, and aerospace would be especially well-suited to this work.

Civilisation Resilience

Making societies more robust to the dangers of a nuclear exchange could be an effective strategy to mitigate the risks of nuclear war. This includes projects such as building refuges and bunkers, as well as widespread changes like making food systems more resilient.

This may be a more accessible route to impact for some engineers. Building civilisational resilience is a burgeoning field and may require a different background from many of the other career paths to reducing nuclear risk. Civil engineers, biomedical engineers, chemical engineers, those with experience in the agricultural industry and others could all contribute here but may find many of the other bottlenecks difficult to pursue.

Becoming an expert in nuclear testing, modelling, and simulation

In the wake of various test ban treaties, there was a necessity for many nations to develop new methods to conduct experiments on nuclear arms systems. Today, nuclear powers without testing campaigns conduct lengthy and expensive experimental research to verify and modernise their arsenals.

Notable in this is the development of extremely advanced modelling and simulation tools to analyse many aspects of these weapon systems. Many of the world’s most advanced supercomputers are used almost exclusively for these tasks.

Expertise in this modern age of testing may be a key bottleneck for policymakers. The topic is extremely complicated and some efforts to put controls on this space have been poorly implemented. Engineers with experience in modelling and simulation or high-performance computing would be especially well-suited to this work.


The MacArthur Foundation was the largest private funder of nuclear security work with grants of around $15 million per year between 2014 and 2020, around 32% of the total annual philanthropic nuclear security funding. In 2021, they announced that they would withdraw funding support, with final funds disbursed in 2023. This leaves a major funding shortfall. For comparison, the budget of Oppenheimer was more than twice as much as philanthropists currently spend on preventing nuclear war.

With tensions building due to the Russia-Ukraine war and other conflicts, donations now could have particularly high leverage over the strategic direction of the field.

Career moves

Positioning yourself as a policy expert

As many of the interventions to reduce the risks of nuclear war are political actions, policy experts are very well positioned to have a high impact.

For some engineers, returning to education to get a graduate degree in a relevant subject such as international relations or public policy may be beneficial. This can make it more straightforward to get into defence policy, but it is also possible to enter into the field without these qualifications.

Technical experience is valued in the field, so having years of experience in engineering before beginning work on policy is beneficial. It would make sense for many engineers to target an area of policy that is quite technical (such as emerging technologies), as they may have more credibility here.

Nuclear weapons policy touches on many technical areas. Becoming an expert in the intersection between nuclear weapons and an emerging technology which which one has existing expertise is an effective strategy for building credibility. This would include the intersection between nuclear weapons policy and topics such as AI, nanotechnology, and computational modelling.

Particularly high-impact areas of policy include:

  • Disarmament advocacy: Disarmament has led to a huge reduction in the potential death toll from nuclear war and remains the most effective way to lower nuclear risks.
  • Targeting strategy: Changes to a nation’s targeting strategy could have a huge impact on the death toll in a nuclear exchange. Presently, both military infrastructure and densely populated cities would be targeted by most nations in a nuclear strike. Among the justifications for bombing cities is that they contain political centres, manufacturing facilities and key infrastructure. Having targeting plans target fewer cities could save millions of lives in an exchange.
  • Preventing accidents: More strict policies preventing accidents, thefts and improving security at nuclear facilities and in command and control systems would be very impactful.
  • Non-proliferation: Proliferation of nuclear material and nuclear weapons destabilises deterrence and international cooperation.

One notable program for those interested in transitioning to policy is the Princeton School on Science and Global Security, which trains scientists and engineers from around the world in technical perspectives on understanding, reducing, and ending the threat from nuclear weapons. It is a short program that runs each year and mostly accepts graduate students in STEM.

Other policy fellowships include the Science and Technology Policy Fellowship at the American Association for the Advancement of Science and the Stanton Nuclear Security Fellows Program at RAND. Further policy fellowships can be found on this database of US policy fellowships and this guide to the European Commission’s Blue Book Traineeship.

Studying at graduate schools with an interest in averting nuclear war such as the Middlebury Institute of International Studies,  The Alva Myrdal Centre for Nuclear Disarmament in Uppsala University Sweden, the Vienna Center for Disarmament and Non-Proliferation, the Centre for Nuclear Non-Proliferation and Disarmament at the Australian National University, The Simons Centre for Disarmament and Non-Proliferation at UBC Canada, and the Centre for Science & Security Studies at Kings College London would be an excellent first step.

Positioning yourself as a defence or arms technology expert

Another route to involvement in a nation’s nuclear strategy is through being involved in a nation’s military, intelligence agencies or defence companies. These organisations can have an effect on nuclear strategy in most nations. This strategy should only be attempted by those who are already unusually well-positioned to do so. The agency of individuals is generally limited in these institutions.

Engineers who already have experience in defence technologies may be well-positioned for this.

Risks, pitfalls, and things to keep in mind

Much of the work within the field of nuclear weapons is likely either actively harmful or is extremely uncertain if it would have a positive effect. Engineers are especially vulnerable to doing harmful work in this field despite good intentions, as it is often assumed that engineers are happy to work on any technically challenging project regardless of its positive or negative impact. This assumption is not poorly founded, as many are. This means that an engineer’s ethical or political position is often not taken into account during hiring or when an organisation is putting teams together for projects. For example, this article’s author had completed several recruitment steps for a research project, unaware the research was funded by a missile manufacturer to design more effective missiles, some with nuclear capabilities.

With this in mind, we recommend avoiding the following notable areas:

Avoid improving missile defence systems

Intercontinental ballistic missile defence systems can seem on their face to be a viable strategy to avert or reduce the damages incurred in a nuclear exchange. However, their testing record and the international response to their development tell a different story.

Missile defence systems have proven in full-scale testing to have a very limited success rate at intercepting intercontinental ballistic missiles. This is true even in a best-case scenario where the oncoming missile is not equipped to overcome missile defence systems and the trajectory and timing of the missile are known.

The development of missile defence systems has proven to be relatively slow and extremely expensive. For countries pursuing these systems, such as the US, this has led to adversaries building advanced delivery systems such as multiple independently targetable re-entry vehicles, fractional orbital bombardment systems, hypersonic glide vehicles and other technologies designed to overcome missile defences. These technologies have proven cheaper and easier to develop and produce than missile defence systems. In a future world with less stable disarmament norms, a country could also merely increase its nuclear arsenal to overwhelm these defences at a low cost.

Intercontinental ballistic missile defence systems are ineffective and an extremely destabilising force to deterrence and disarmament norms. Work related to advancing these systems should be avoided and great care should be taken in any work that might be related to missile defence.

It should be noted that the smaller “theatre” missile defence systems may not carry the same implications. These intercept missiles with a range of less than 3,500 kilometres, often within combat zones. These have less of an effect on the deterrence strategy in place for ICBM use and may be helpful in reducing battlefield deaths.

Avoid advancing the capabilities of weapons systems

As one might expect, advances in nuclear weapons systems will typically make them more deadly and increase the destruction in a nuclear exchange. This includes advancements in yield, speed, range, manoeuvrability, telemetry and overcoming missile defences. As technical professions in this field have very little strategic input, engineers should not attempt to change organisations such as defence contractors from the inside.

Significant advances in weapons systems generally also lead to arms races, where either the same technology or a negating technology is rapidly developed by adversaries. Therefore, even if there was a nation that the whole international community would benefit from having weapons superiority, it may not be wise for that nation to pursue advancements. Incidentally, it is not at all clear that such a nation exists today.

Advancing weapon accuracy is one notable exception. Greater accuracy could lead to more effective targeting of military infrastructure and therefore fewer casualties due to smaller weapon numbers and yields being used to hit targets. However, attempting to advance weapon accuracy is likely a suboptimal strategy for impact.

Engineers should not pursue careers in advancing weapons systems and instead work on interventions that plausibly directly reduce nuclear risks. If technical expertise in aerospace or nuclear physics is required for some specific career path, these skills are often more easily obtained in the non-military aerospace or nuclear power industries respectively.

Avoid spreading information hazards

Information hazards are defined as a risk that arises from the dissemination or the potential dissemination of information that may cause harm or enable some agent to cause harm. Publishing technical information, drawing attention to sensitive data sources, or releasing information that inadvertently signals the intentions of a nation to others could all increase nuclear risks under certain circumstances.

Information hazards are detrimental to non-proliferation efforts, as they could aid nations or even terror groups in developing nuclear capabilities. Under some circumstances, they can also be detrimental to international relations.

The open source intelligence community may be an especially concerning source of information hazards. Although much of their work is beneficial to transparency, there is potential to release information to the public that causes great harm and extreme caution should be taken by those doing openly published work.

In order to work in this field some amount of sensitive information needs to be handled and communicated to others and extreme care should be taken in this. Work in this field is however extremely important and should not be disregarded for fear of information hazards alone.

Avoid advancing radiological weapons or their underlying technologies

Radiological weapons disperse radioactive material to inflict radiation damage or cause radiation contamination of an area. Though these weapons are often thought of in relation to non-state actors or terror groups, they have been developed by several nations in the past.

Radiological weapons are of special concern as they can produce a much more intense fallout than regular nuclear weapons and can render an area uninhabitable for a long period. The radiation from a salted bomb for example would initially be 35 times more intense than a typical fission bomb and would only reduce to the same level as the fission bomb after 75 years due to its shorter radiation half-life.

The long-lasting effects of radiological weapons and the potential spread of radioactive material present a serious and more long-lasting threat to humanity if used and work on them should be avoided. Thankfully developing weapons employing radiological warfare is uncommon today, though some development has been seen in recent years.

Despite the high potential impact, nuclear safety is likely one of the highest risk areas to do technical work, similar to technical work in Biorisk and Biosecurity. Any attempts at impact in this cause area should be done with careful contemplation and consultation with others in the field.

Learn more

Additional resources

Relevant organisations

We are grateful to MA for their input and feedback. All opinions and remaining mistakes are our own.

Sorted by Click to highlight new comments since:

This is some great research and a really nice summary. I read it on the High-Impact Engineers Portal, but great to see it being cross-posted here, as it's just a very good summary of the state of nuclear risk in general, and the challenges we face. 

It's great how you highlight the real risks. I see a lot of analogy to AI here; you can go into this field with the best of intentions, and if you're not careful, you could end up working on something that is increasing rather than reducing the risk. 

Thanks for doing and writing up this research!


Thanks Denis, glad it was helpful/interesting for you! 

Thank you for this extensive overview! Definitely relevant and I'll be sure to keep it for my coachees who want to switch careers into nuclear-related paths. Do you have any knowledge of a repertory of experts working in these institutions that could be contacted by my coachees in case of questions related to these jobs? 

Thank you for your positive feedback! I am not aware of a particular directory of these experts, but I'd be happy to connect you to our contacts working on reducing nuclear risks.

Curated and popular this week
Relevant opportunities