TY - JOUR
AU - Dixon, Thom
AB - Abstract The convergence of the life sciences with the information and computing sciences is beginning to generate novel security vulnerabilities. As scientific and technological advances occur, new security vulnerabilities are discovered and new methods for exploiting those vulnerabilities are developed. Novel cyber-biological capabilities are likely to enable technologically sophisticated states to develop new methods of grey zone warfare. This article provides context to this multidisciplinary area of research by reviewing the emerging field of cyberbiosecurity for its relevance to developments in grey zone warfare. This article then analyses two long-term trends that have influenced the development of contemporary cyber-biological capabilities. These two trends are advances in novel uses of biology and advances in computing, automation and biodesign. The capability to exploit vulnerabilities unique to the links between cyber and biological systems differs significantly from previous security concerns noted for biotechnology and the life sciences. Scholars and practitioners of international relations will need to develop an understanding of engineering biology and the bioeconomy in order to forecast methods of grey zone manoeuvre that rely on cyber-biological capabilities. This article offers an entry point for the scholar and practitioner so that they may bring their own disciplinary lens to the issue of grey zone ambiguity and cyberbiosecurity. On 17 September 2019, the US Bipartisan Commission on Biodefense convened a study panel entitled ‘Cyberbio convergence: characterizing the multiplicative threat’.1 The topics covered included security vulnerabilities arising from the convergence of the cyber and biological sciences, the vulnerability of pathogen and biomanufacturing systems, biological risk mitigation, and the vulnerability of intellectual property in the national and global bioeconomy. This article pursues that panel's concerns by highlighting how cyber-biological capabilities enable grey zone warfare strategies, and by analysing the scientific and technological forces that underpin these emerging capabilities. Advances in the life sciences have made biological systems easier to engineer at ever increasing speeds and ever decreasing cost.2 These capabilities enable the sophisticated adversary to exploit novel vulnerabilities unique to biological systems, biological information, and the links between cyber and biological systems.3 These emerging capabilities are dual use in the sense that they can be deliberately misused.4 In this article I consider those uses that benefit the economy, society and human health, while also enabling deployments for defence, security and intelligence advantage. As the article will show, the ability to exploit an adversary's cyber-biological vulnerabilities enables technologically advanced states to undertake sophisticated grey zone warfare strategies.5 These strategies exploit grey zone ambiguities that enable camouflaged operations with plausible deniability.6 In 2010, Professor Gregory Koblentz identified four trends contributing to international biosecurity challenges: advances in science and technology; the emergence of new diseases; globalization; and the changing nature of conflict.7 This article will focus on the first and fourth of these—advances in the science and technology of engineering biology, and the changing nature of conflict. The increase in grey zone conflict between Great Powers is a key reason why advances in engineering biology are relevant to international security. That is, emerging cyber-biological capabilities are likely to enable novel methods of plausibly deniable grey zone manoeuvring. This article is structured in four sections: (1) a discussion of grey zone warfare and cyberbiosecurity; (2) an analysis of recent advances in novel uses of biology; (3) an analysis of recent advances in computing, automation and biodesign; and (4) a conclusion discussing the ambiguity of novel cyber-biological vulnerabilities. It argues that security scholars and practitioners need to re-imagine the capabilities of biology. Three examples of current research exemplify this need: (1) a project funded by the Defense and Advanced Research Projects Agency (DARPA) seeking to identify submarines by characterizing the signal patterns of the living ocean biome in relation to submersible proximity;8 (2) research into sentinel plants that enable the persistent sensing and surveillance of chemicals, compounds and fragrances in the local environment;9 and (3) research into the integration of engineered bacteria into soft robotics as environmental sensors.10 The cyber interfacing of biological information from natural and engineered systems is different from previous security concerns noted in connection with biotechnology.11 This confluence of information and biology enables technologically sophisticated actors to exploit a proliferation of unique biological vulnerabilities. This article is intended to communicate developments in the life sciences to an International Relations (IR) readership. Where possible, scientific terms and jargon have been simplified, and the terms ‘living systems’ and ‘non-living systems’ are used in describing information security issues arising from the interfacing of animate and inanimate information substrates. Cyberbiosecurity and grey zone warfare Cyberbiosecurity is a term that refers primarily to the cyber security of biological information, whether this be digitally stored genomic information, its descriptive metadata, scientific research, or health care and medical information.12 It has been defined by the co-originators of the concept as understanding the vulnerabilities to unwanted surveillance, intrusions, and malicious and harmful activities which can occur within or at the interfaces of comingled life and medical sciences, cyber, cyber-physical, supply chain and infrastructure systems, and developing and instituting measures to prevent, protect against, mitigate, investigate and attribute such threats as it pertains to security, competitiveness and resilience.13 This definition should capture the security issues arising from the novel cyber-biological systems mentioned above, but in practice cyberbiosecurity has primarily focused on the vulnerabilities of biological information when stored in non-living digital systems. The 2019 study panel referred to above on ‘Cyberbio convergence: characterizing the multiplicative threat’ is a good example of this focus.14 The long-term usefulness of cyberbiosecurity will be determined by its ability to assist in the assessment and mitigation of novel information security issues regardless of whether the information comes from a living system such as a sentinel plant or a non-living system such as a cloud computing data centre. A commentary published in 2012 after the H5N1 flu outbreak describes how cyber-security methodologies could be applied to the biological domain.15 Since then, the practice of cyber security has developed increasingly sophisticated approaches to information security, but these techniques are often specific to digitally structured information. As the life sciences and the information sciences continue to converge, sophisticated cyber-security techniques will need to be adapted for novel integrations of living and non-living systems. The integration of biological and digital systems will challenge traditional security constructs because it will challenge traditional understandings of information. Cyberbiosecurity practitioners will need new skill sets in order to secure living-system information within living-system information substrates, such as chemical signals in a sentinel plant,16 bioelectrical signals interfacing living and non-living systems,17 or digitally structured information written to DNA storage.18 Cyber attacks target the most vulnerable point in an information technology system. As the interfacing of living and non-living systems becomes more widespread, it will be important for cyber- and bio-engineers to ensure that living components are not the most vulnerable point in the information management of an overall system or process. The addition of the terms ‘cyber’ or ‘bio’ only provides a reference point regarding the context of a security vulnerability, and over time the accuracy of these reference points may degrade. The co-originators of cyberbiosecurity in part acknowledge that their taxonomy will come under strain, and they call for the creation of a more robust taxonomy in their 2018 article.19 With this in mind, I propose a revised definition of cyberbiosecurity that is inclusive of contemporary research in cyber-biological interfacing: cyberbiosecurity encompasses those biological, medical and genomic information security vulnerabilities that arise from the interfacing of living and non-living systems, and the integration of living (animate) and non-living (inanimate) information substrates. This definition is intended to be inclusive of all types of biologically descriptive data and metadata, including domain-sized biome behaviour signals. It is also intended to be inclusive of all forms of information storage and communication common to living and non-living systems, for example, electronic, chemical and optical substrates,20 but not exclusive of more exotic information storage and communication substrates relevant to cyberbiosecurity systems, such as quantum biology, sonics and bioelectrochemistry.21 A key critique of this definition is that it is too broad to be useful. However, this article proposes that there is a need to group holistically the different ways in which information about and from living systems (including humans) is interfaced with and stored in digital systems. From a cyber-security point of view there may be very little difference between the storage and interfacing of living system information with information from non-living cyber-physical systems. From a biosecurity point of view, however, the process of interfacing living and non-living systems and co-mingling their information has fundamentally changed the practice of engineering biology, opened up entirely new biological design spaces and created novel biosecurity issues. When the issue is viewed from a biosecurity vantage point that takes into consideration the past two decades of transformative advances in the life sciences, it is clear that an inclusive definition for cyberbiosecurity will be necessary if future security issues are to be adequately addressed. This need has been captured by an article on the future of bio-informational engineering that proposes the re-imagining of natural and engineered biological systems as cyber-physical architectures.22 Security issues arising from advances in areas such as bio-informational engineering are unlikely to be mitigated by technical solutions without supporting political solutions. As Sara Davies and Clare Wenham have noted in relation to COVID-19 and the World Health Organization,23 devising solutions to these security issues will require the discipline of IR to collaborate with scientists and life sciences practitioners. Living systems remain the only systems that can replicate themselves in the real world. The capability to digitally actuate biological functionality24 with high specificity in time and space and with no off-target effects represents a unique development in the history of science and technology. It is a capability distinctly different from that of cyber-physical systems that, for example, interface chemicals, electronics or robotics with digital systems. Neither chemicals, electronics nor robotics can currently build themselves, replicate in the wild, or evolve the functionality of their physical manifestation with no human intervention. Living systems are unique in this capability, and the advance of science in this area necessitates an inclusive and holistic cyberbiosecurity definition. This article is not the first to propose the cyberbiosecurity concept, which has been under discussion in scientific journals since 2018.25 It is, however, the first to propose the concept to an IR readership and the first to articulate why the discipline of IR needs to be aware of its development. Arms control regimes Arms control regimes such as the Convention on the Prohibition of the Development, Production and Stockpiling of Bacteriological (Biological) and Toxin Weapons and on their Destruction (Biological Weapons Convention: BWC) and the Chemical Weapons Convention (CWC) are likely to come under strain because they will find it difficult to maintain parity with an increasing array of plausibly deniable uses for exploiting biological vulnerabilities. Similar challenges will arise for UN Security Council Resolution 1540 (regarding weapons of mass destruction) and its subsequent iterations.26 Across the coming decades, the BWC will continually need to address the advance of science and technology in its intersessional programme, as well as its review conferences.27 Similarly, discussions of article VII of the BWC (which requires state parties to assist other states that have been exposed to danger as a result of a violation of the convention) are likely to become increasingly important in the future. Even though article VII has never been invoked, grey zone activities deploying offensive cyber-biological capabilities could trigger this article in the future. The status of article VII in relation to cyber-biological activity may require clarification by BWC state parties. There is, however, likely to be an inherent issue with the efficacy of the BWC and grey zone use of offensive cyber-biological capabilities. State parties agree in article I not to develop, produce, stockpile or retain weapons, equipment or means of delivery designed to use such agents or toxins for hostile purposes or in armed conflict; but grey zone warfare necessarily relies on ambiguity around the definition of ‘hostile purposes’ as well as intentionally clouding the attribution of use. Defining the grey zone The reference points of cyber security and biosecurity are important for forecasting advances in the methods and technologies of grey zone warfare. As David Kilcullen notes, a grey zone is identifiable because it has an ambiguous political, legal and psychological status.28 Grey zones are characterized not by concrete borders but rather by the existence of ambiguous transitional zones. Grey zone warfare exploits ambiguity as a form of manoeuvre that is neither fully overt nor truly clandestine. As a tactic and a strategy it exists at the edge of detectability, characterized by the exploitation of undefined or legally ambiguous spaces and categories. By acting through these spaces, grey zone warfare creates camouflage for action without retaliation—plausible deniability. Colonel Gary P. Corn has written on the grey zone in relation to cyberspace and suggests that it can be thought of as a distinct operating environment existing on a continuum of conflict somewhere between peacetime and overt warfare.29 However, Corn notes that grey zone threats are manifest across all domains, even though cyberspace is a particularly lucrative medium. The grey zone was formally defined by the US Office of the Secretary of Defense in 2016 as a conceptual space between peace and war, occurring when actors purposefully use multiple elements of power to achieve political–security objectives with activities that are ambiguous or cloud attribution and exceed the threshold of ordinary competition, yet fall below the level of large-scale direct military conflict, and threaten US and allied interests by challenging, under-mining, or violating international customs, norms, or laws.30 Though this definition is specific to the US context, it highlights existing work on the application of the grey zone to cyber security that can be repurposed for a cyberbiosecurity assessment. As Mark Hoffman and Martin Hofmann have noted, the grey zone can be a significantly more complex operating environment than cyberspace because cyberspace is just one domain within the broader grey zone environment.31 It is because of the multi-domain nature of grey zone warfare that an appreciation of cyberbiosecurity is important for anticipating future strategic and tactical developments in grey zone manoeuvre. Cyber-biological interfacing is neither a biology-only nor a cyber-only exercise, but rather occurs at the margins of these two broad disciplines. This means that cyberbiosecurity capabilities are transitional and ambiguous. As such, they are increasingly likely to sit outside, or at the very edge of, the traditional reach of international arms control mechanisms such as the BWC, the CWC and UNSC 1540. These mechanisms rely on bounded disciplinary definitions like those demarcating the biological sciences and the chemical sciences. Technologically sophisticated states will be able to exploit these artificial boundaries by confining their exploitation of novel biological vulnerabilities to plausibly deniable uses. For example, as the capacity to synthesize DNA improves, sophisticated actors may accrue the ability to spoof DNA at a crime scene.32 This will allow an actor to camouflage the true perpetrators of a crime while erroneously pointing law enforcement at an innocent target. For such a hypothetical scenario to work, the malicious actor would necessarily have had to acquire either a digital or a biological instance of the target's DNA prior to the operation. As another example, sophisticated actors could develop bioweapons that mimic the symptoms of common diseases in order to camouflage the initial spread of an engineered pathogen.33 This would require technologically advanced cyber-biological capabilities and human resources. While this application falls explicitly within the domain of the BWC, in practice it may be very difficult for states targeted with such a weapon to attribute use in a timely manner without their own technologically sophisticated defensive cyber-biological capabilities. As a final example, current research in cyber-biological interfacing is opening up the long-term possibility of signal-activated biological devices.34 While such technologies may be game-changing for precision medicine, they represent a potentially significant dual-use research development. Indeed, such a capability could alter the offensive–defensive balance for biological weapons by enabling research into the signal activation of pathogenic traits.35 Such research could feasibly enable research into bioweapons with signal-based targeting and triggering capabilities that achieve camouflaged action by mimicking common disease symptoms. It will be important for states to work closely with their scientific communities to ensure they are correctly assessing the technological readiness of adversary states in relation to such capabilities.36 A diverse range of IR expertise will be required to devise solutions for these issues, and this article was written to highlight that need. Such work will be best placed within existing frameworks of responsible research and innovation and collaborative research into the ethical, legal and social implications of life sciences research.37 Cyber-biological convergence is creating gaps in the coverage of existing arms control regimes when it comes to the exploitation of biological vulnerabilities. These gaps are likely to grow as science advances its understanding of biological systems ranging from enzymes, single cells, microbiomes, plants, animals, humans and environmental biomes through to the various and many types of cyber-physical systems. Over time, this may encourage the development of novel offensive capabilities that deploy weaponized biological or cyber-physical systems, or target the information transfer and storage mechanisms that connect living and non-living systems. An emerging vulnerability likely to be exploited across the next decade will be the use of illicitly acquired genomic information and medical data as a pathway to target selection for intelligence operations. Advances in social genomics suggest that traits such as loyalty and addictiveness may soon be correlated with genetic and epigenetic patterns harvested from a target's cyber-biological profile.38 If such traits have marginally significant statistical correlations with an individual's cyber-biological profile, they are likely to be of profiling interest to intelligence agencies weighing up recruitment targets. Grey zone warfare often involves initially camouflaged forms of manoeuvre,39 intended to enable gains to be consolidated before the adversary can attribute the attack and respond. In these scenarios, a grey zone warfare strategy can be enhanced by the exploitation of an adversary's biological vulnerabilities. Tactics that confuse the defender's command structures elongate the window of time in which an offensive force can manoeuvre freely and consolidate tactical gains into strategic wins. Strategies that support this objective could include targeting the biological vulnerabilities of key decision-makers in an adversary's command structures.40 The medical and genomic information of military and civilian personnel is likely to be an essential input for any future strategy of grey zone warfare involving targeted incapacitation or assassination under the cover of natural causes.41 There is a need for more research into the offensive uses of cyber-biological capabilities and how such capabilities can augment or amplify existing offensive uses in biological or chemical weapons. Cyber-biological capabilities that disrupt disease surveillance systems, compromise medical response systems or attack vaccine manufacturing supply chains could each be more effective than command structure disruption, especially when combined in a layered manner. Each of the capabilities noted above deserves research into early warning and mitigation mechanisms within the context of grey zone warfare. Indeed, there is much work to be done on a host of new security questions exploring deterrence and the proportionate use of force in relation to cyber-biological grey zone warfare, with a particular focus needed on securing pandemic response vaccination programmes. The potential defensive and offensive uses of genomic, medical and health care data are currently difficult to quantify. There is a growing awareness of this vulnerability among states, evidenced by new laws that either secure or exploit genomic information on a domestic basis. In 2017, China introduced new data protection provisions relevant to medical data, and Beijing has also implemented state-wide DNA surveillance programmes.42 The United States has also enabled law enforcement agencies to access the DNA databases of companies such as Ancestry.com and 23andMe,43 and this approach is estimated to capture up to 60 per cent of the US population of European descent by allowing the identification of relatives (up to and including third cousins) from currently existing genome data.44 Importantly, the US approach to DNA-based identification is believed to be more accurate than the Chinese approach.45 Both approaches seek to make use of biological vulnerabilities arising from cyber-biological interfacing and exemplify the current ethical, legal and social norms in biotechnology. The use of genomic, medical and health care data for surveillance and law enforcement is well under way, and this already enables micro-targeting through biological discrimination at the genomic level. This extension of state-based surveillance into a growing intrusion on personal privacy represents a continuation of long-running trends and may make rational sense from the perspective of the state. However, the cyber-biological nature of emerging surveillance regimes is likely to be viewed by some of the public as ethically and legally problematic.46 Many scholars and practitioners have been anticipating the disruptive potential of engineering biology for some time. The UK-based Centre for the Study of Existential Risk has undertaken two horizon-scanning surveys,47 the US Bipartisan Commission on Biodefense has—as noted above—investigated cyberbiological convergence,48 and much work has been done preparing papers on the advance of science and technology for BWC meetings of experts.49 Engineering biology promises to be a general-purpose scientific technology akin to semi-conductors and optoelectronic engineering. Early indications are that engineering biology may have impacts on every sector of the economy and society over the coming decades. This is due to its wide-ranging applicability to a number of grand challenges relevant to twenty-first-century non-traditional security issues such as climate change mitigation and pandemic preparedness, as well as to sustainable circular economy models of development in the areas of medicine, public health, energy, materials and manufacturing.50 There is, therefore, a need for a much wider grouping of IR scholars to engage with the emerging world of engineering biology and the bioeconomy. With this in mind, the next three sections of this article chart the structural forces of change contributing to the accelerating advance of cyber-biological engineering and interfacing. It is of course important to note that all these forces are associated with positive economic, social and environmental impacts,51 and these positive factors are a key force underpinning the continuing advance of the relevant science and technology. Advances in novel uses of biology This section traces the recent history of novel uses specific to synthetic biology that form a critical component of engineering biology.52 Each novel use has expanded the design space of bio-engineering, in turn enabling new discoveries and new uses of biology.53 Biosensors are a recent development in the history of novel biological discoveries and inventions. Biosensors are biological devices that use programmed metabolic pathways to monitor and report on changes in their internal and external environments.54 These devices are engineered to transform biological inputs and processes into macroscopic informational outputs.55 These outputs are often conveyed by engineering a fluorescent or bioluminescent output signal into a biological device.56 In laboratory settings these biological devices have been used to speed up the biodesign process, allowing for more rapid identification of successful organism strains.57 As biological processes become better understood, this significantly accelerates biodesign timelines.58 Outside laboratory environments, biosensors are useful in solving a range of problems in civilian and defence scenarios. For example, biosensors can be used to identify trace amounts of chemical compounds in the environment, and this has implications for monitoring chemical and biological weapons. They can convey real-time environmental information through simple mechanisms like bioluminescence. Biosensors translate dynamic environmental attributes into usable real-time intelligence, and integrating biosensor data into digital networks represents the next step for the automation of engineering biology. Digitally connected biosensors create opportunities for persistently monitoring the environment at the micro-biological scale, in turn enabling unique intelligence collection channels for harvesting natural and engineered biodata. Biosensors can also be deployed for a large range of commercial applications,59 and this potential is likely to generate commercial tools for the big data information management of biome-level sensing and surveillance networks. In 2019, a research team developed a soft robotic gripper that deployed engineered E. coli to detect chemicals on the objects with which the gripper interacted.60 The gripper integrated a luminescent biosensor with an optical sensor, creating an autonomous robot that could respond to chemical stimuli. This is an example of a system integrating different control loops across multiple engineering areas, drawing together chemical, optical, electrical, digital and mechanical processes. It indicates the range of functionality engineered systems can achieve when combined with biosensor functionality. This type of cyber-biological interfacing is in its infancy, but the commercial deployment of these systems will impact many industries, particularly those in the defence sector. One early application will be the production of advanced materials. Research in engineering biology was initially confined to biomimicry and materials design.61 For example, spider silk has a tensile strength five times as strong as the same weight of steel and three times as strong as Kevlar.62 Basic research into spider silk has shown that the way a spider spools its web is critical to ensuring this high tensile strength. The initial chemical composition of the silk is relatively weak, and a novel approach is required to synthesize this material in a way that captures the high-tensile benefits.63 This is important for those seeking to chemically synthesize enough spider silk to manufacture clothing and equipment with better defensive properties than Kevlar. But as the soft robotic gripper example demonstrates, engineering biology is now solving the problem of information communication between biological (chemical, electrical and optical) and digital (electrical and optical) systems. The integration of biosensor data into satellite networks and internet of things telecommunications will change key assumptions regarding sensing in both the built and natural environments. Advances in living monitoring systems could realize microbiological-level surveillance networks that communicate in real time via satellite link. These kinds of capabilities could open novel vectors for spoofing, hacking and compromising digital and biological systems—for example, side channel attacks that capture commercially sensitive or classified intelligence via scaped emissions of bio-information.64 These capabilities create novel vectors for intelligence collection, the monitoring of emerging infectious diseases and the automation of agriculture. These systems extend the science and technology underpinning engineering biology into new design spaces. Though these capabilities are in their infancy, they could be used to deploy real-time control loops that trigger drone surveillance and targeting, or the repointing of high-resolution satellites and the delivery of kinetic targeting solutions. During the next decade, it is likely that this will lead to the integration of biological informational inputs into a range of human-designed living monitoring systems. More importantly, it is easy to imagine a diversity of ways in which such capabilities could be used to enhance automated anti-access and area denial systems that could be theatre-wide in scope or enable persistent environmental surveillance of large swaths of the maritime and terrestrial domains. In the short term, next-generation cyber-biological interfacing is likely to be refined through use in scientific laboratories and industrial applications. This is coming to be known as a ‘full stack’ approach to engineering biology,65 and is supported by a growing commercial community. For example, biological design and bioinformatic software solutions are serviced by companies such as Benchling, Synthace (through their Antha software), Ryffin and Teselagen. Then there are biology tool component services such as Synthego and Caribou. Other companies focus on increasing product yield and addressing scale-up challenges. Still others form vertically integrated ‘stacks’ or horizontal ‘platforms’ that bundle services together to target specific markets or consolidate work across industrial sectors. Ginkgo Bioworks, BGI and Twist Bioscience are good examples of this. Genome foundries, or biofoundries, deploy high-throughput technologies within automated and modular workflows, enabling a full stack approach to biological engineering. In doing so, they vastly increase the speed and scale at which the engineering of biology can be achieved. Biofoundries are already creating artificial intelligence–enabled workflows.66 By deploying reinforcement learning and deep-learning techniques, organisms with desired phenotypic traits can be identified earlier in their growth phase—prior to when human observation can successfully distinguish desired macroscopic candidates. This builds on decades of work in bioinformatic image recognition,67 and can further reduce the time between design cycles. Machine learning applications are lowering the effort required to iterate through generations of designs, and therefore reducing the time required to design a biological solution for a given problem set. Cyber-biological interfacing is bringing about a phase change in engineering biology, building on the novel capabilities of biosensor functionality that have emerged from over two decades of advances in basic and applied biological research. The discipline is now creating biological devices and control systems that replace or integrate traditional chemical, mechanical and digital workflows. This has created entirely new industries and opened up new domains of bio-economic opportunity. However, these technologies are dual use. Not only do these advances bring their own unique vulnerabilities, they also enable the design of novel dual-use biological devices. This increasing novelty has in part arisen from the supporting infrastructure and equipment, and the coalescence of multidisciplinary expertise relevant to engineering biology. It is, therefore, time to turn to another trend that underpins much of engineering biology's contemporary capabilities. This trend has been driving advances in the science and technology of engineering biology for many decades, but its contributions have become particularly acute during the past 20 years. Advances in computing, automation and biodesign Both Associate Professor Gigi Kwik Gronvall at the John Hopkins Centre for Health Security and NASEM, in its 2018 report Biodefense in the age of synthetic biology, acknowledge that advances in computing have probably been the greatest enabler for research and development in the field of synthetic biology.68 This section contextualizes how advances in computing and automation have contributed to novel dual use in engineering biology. Semiconductors and optoelectronics form the backbone of computing and telecommunications.69 They have given rise to a range of life sciences software programming languages for computer-assisted design (CAD) that elevate the abstraction of biology to ever higher degrees.70 Techniques similar to those used in much better-known forms of computational abstraction (for example, interacting with binary information through a word-processor application) are used for the four-base chemical language of DNA. These software programs abstract genomic information into that of genetic parts and genetic circuits. These parts are then functionally characterized through experimentation and the resulting metadata are catalogued in publicly accessible, commercially confidential and state-classified biodefence databanks.71 These data then become a digital feedstock enabling the abstraction of biology to ever higher degrees. Importantly, open-source databanks include the complete genetic sequences of many pathogens, including smallpox, anthrax and Ebola haemorrhagic fever.72 Monitoring the use of open-source digital instances of pathogens may become more important as the resource and knowledge thresholds for engineering biology continue to decline. Indeed, there are CAD software solutions today that conduct design driven by artificial intelligence.73 It is important to ensure these software solutions are not vulnerable to deliberate or accidental misuse. Over the past decade, the life sciences have taken the transformative trends inherent in semiconductor and optoelectronic advances and adapted them to enable the information-managed engineering of biology. The enabling capability of genomic sequencing and synthesis is only as great as the supporting software for each level of abstraction. In 2010, the University of California at Berkeley hosted the inaugural Critical Assessment of Genome Interpretation (CAGI) competition.74 From these beginnings the genomic design, build, test and learn (DBTL) software market has grown, making use of modern conveniences such as cloud computing, browser-based applications and machine learning. This is creating a world in which the end-user constraints on engineering biology are primarily knowledge-based. Through cloud computing infrastructure, organism design has been divorced from the need to access high-performance computing. This means anyone with threshold-level knowledge can subscribe to a biodesign software solution and order their novel genetic parts from a commercial synthesis provider. State and non-state actors can then hire laboratory space to validate their designs and need not have a commercial or organizational footprint of the kind traditionally associated with the ‘big science’ of synthetic biology. Current US dual-use regulation does not capture dual-use research of concern (DURC) among commercial organizations that own and undertake experiments in laboratory space outside the United States. For example, the synthesis of horsepox75—and the open-source publication of the method—is an example of DURC in engineering biology occurring in a laboratory in Canada that was not covered by existing US regulation.76 Given that US dual-use regulation tends only to apply to federally funded research, non-academic organizations with no institutional footprint may pose a challenge to the efficacy of existing regulatory regimes. The monitoring and regulating of engineering biology DURC in a global context represents a greater challenge. This is because no one country is the sole supplier of biodesign software, DNA synthesis technology or laboratory rental space. Meanwhile, commercial DNA synthesis providers (better known as biofoundries) are finding that much of their corporate value is bound up in their proprietary algorithms and proprietary software. The poster child for this full-stack revolution in biodesign is Ginkgo Bioworks. In 2019, Ginkgo began offering its proprietary biodesign platform to new start-ups through its venture capital programme. Microsoft's Station B and Teselagen's software platform are both examples of engineering biology CAD enabled by artificial intelligence. Codexis and Zymergen already use their proprietary software and artificial intelligence capabilities to design novel proteins for desired functionality, with applications in agriculture, electronics and sustainable materials. All this software-enabled activity is underpinned by a growing pool of data describing the design space of life—or, more importantly, data that describes the solution space of life.77 These data are becoming an increasingly essential component to scaling up the engineering of biology, and nations that have access to the data can outpace those that don't. As this pool of data grows, so too does the ability of commercial and research biofoundries to offer clients a biological solution based on nothing more than an understanding of their desired feedstock and output. Advances in the supporting software solutions of engineering biology have enabled modularity and comparative advantage to take hold among both upstream and downstream providers in the DBTL stack. Not only has this increased the complexity of the commercial ecosystem involved in sequencing, designing and synthesizing biological devices; it has internationalized it. This commercial web both enables and supports the closely linked basic and applied research of academia. It is through the links between these two systems that the emerging world of biofoundries is developing.78 The long-term trend of increasing performance and decreasing cost that describes semiconductor and optoelectronic manufacturing also holds true for DNA sequencing and synthesis technologies. There is one major difference, though: whereas data storage and processing capacity has tended to follow a ‘Moore's Law’ trend of doubling every year and a half,79 the cost of DNA sequencing decreased at a far greater rate over the period 2002–2012.80 The Human Genome Project cost US$3 billion,81 but by 2010 the sequencing of one human's genome cost US$50,000. Then, in 2014, the company Illumina offered human genome sequencing for US$1,000. Finally in 2020, BGI (formerly known as the Beijing Genomics Institute) announced it would soon be able to offer the sequencing of a human genome for US$100.82 Even if the US$100 rate is aspirational, the historical rates of reduction in cost and advances in technique that characterize the development of DNA-sequencing technology are notable in that they surpass the fastest growth rates in the history of electronics. Genome foundries, or biofoundries, are now displaying similar trends in technological improvement for DNA synthesis to those that have occurred in DNA sequencing. In 1980, DNA synthesis of approximately ten nucleotides cost US$6,000, but by 2010 the synthesis of a million 60-nucleotide oligos83 cost just US$500.84 Biofoundries create exponential improvement curves not just through advancing their equipment, but in the automation of the DBTL workflow itself.85 The combined research and development efforts of both commercial and research biofoundries have a structural impact on the contemporary capacity and capability of engineering biology. Perhaps the most important impact of biofoundries is that the cost of designing, building, testing and scaling up new biological devices is continually decreasing, while at the same time the end-to-end speed of this process is continually accelerating. For an example of this trend at work, one needs only to compare the decade-long biodesign of synthetic artemisinin in 2003 with contemporary full-stack engineering that can process thousands of variants for the same design space in months.86 Not only have the past two decades seen a decrease in end-to-end DBTL processing time, they have also seen a significant widening of the prototyping design space, resulting in more optimal designs at the end of a full-stack iteration. The exploration of much of biology's design space was once prohibited by the experimental expenditure and time required for a full-stack process. As biodesign cost comes down and DBTL performance increases, entirely new design spaces are opened to biological engineering—including, for example, larger and more complex organism groupings (plants, animals and multicellular biomes), novel non-natural compounds (complex chemicals, pharmaceuticals, fragrances, materials and fuels), and nano-scale engineering objectives (DNA nanostructures or DNA origami).87 Meanwhile, machine learning is beginning to enable autonomous biofoundry workflows that iteratively engineer organisms without human intervention.88 Commercial and research biofoundries are only just beginning to make gains in full-stack automation. To put this in perspective, 80 per cent of increases in performance for semiconductor chips are attributed to minor design improvements, while only 20 per cent are attributed to the traditional scaling of transistor size.89 As continual minor improvements occur, they will probably translate into significant improvements in biofoundry performance and decreases in overall biofoundry cost. Improvements in electronic engineering capacity created security issues such as computer viruses, identity theft, privacy invasion and cyberwar. These security issues were emergent characteristics arising from long-term advances in electronic and optoelectronic technologies. Similarly, the international security community should expect that the exponential development of technologies underpinning the engineering of biology will create their own emergent security issues. It should not be surprising that increasing novelty in biological engineering has arisen across the same period of time in which computing hardware, biodesign software and biofoundry automation have experienced an exponential increase in performance and decrease in cost. These two trends are linked, and they will bring with them novel security concerns. Cyber-biological interfacing will open the door to novel grey zone warfare strategies, and the emerging bioeconomy is likely to become the international theatre in which these strategies play out. Grey zone ambiguity and novel biological vulnerabilities These two forces, advances in novel uses of biology and advances in computing, automation and biodesign, have a convergent impact on cyber-biological security. Advances in biotechnology are continually accelerating as the associated costs and resources required of state and non-state actors to use next-generation biotechnology are continually decreasing. This inverse relationship is beginning to show early signs that it will generate a plethora of novel biological vulnerabilities unique to next-generation cyber-biological systems. Cyber-biological interfacing is by definition a liminal scientific and technological practice, and it begets grey zone ambiguities in the same way as do contemporary cyber security and biological security. Detection is difficult, attribution is difficult, mitigation can be costly, and by the time these three facets play out a state or non-state actor is likely to have been able to achieve their original objectives. In 2010, Koblentz identified four trends contributing to biosecurity challenges for international security: advances in science and technology, the emergence of new diseases, globalization and the changing nature of conflict.90 Since 2010, advances in engineering biology have accelerated, technological decoupling between the United States and China has begun, COVID-19 has emerged as a global pandemic, and grey zone warfare has been embedded as a method for states to undertake camouflaged operations with plausible deniability. Biosecurity challenges for international security continue; but, as this article shows, advances in engineering biology fundamentally disrupt key assumptions about what biosecurity threats are and how a state should secure military and civilian assets from them.91 The concept of cyberbiosecurity highlights the likelihood that information security issues linked to living and non-living systems will form a key security vulnerability of the twenty-first century. The range of biological, cyber, information and psychological vulnerabilities that are available for exploitation by the sophisticated adversary is growing. Global geopolitical conditions are deteriorating at the same time as the enabling technologies of grey zone warfare are advancing. Owing to this inverse relationship, grey zone warfare is likely to increase in prevalence over the course of the twenty-first century, and the grey zone is the theatre in which Great Power techno-strategic rivalry will play out. Warfare is changing, technology is changing, and with these changes come new security threats. Never before have biological attributes (human and non-human) been so well understood and so easy to harvest. States need to begin carefully considering where the next cyber-biological attack is coming from and what its intended target may be. Biological information left unsecured today will be worth so much more when it can be used to fuel strategies of grey zone warfare tomorrow. Cyberbiosecurity is in its infancy, and the age of engineering biology has only just begun. Footnotes 1 See the US Bipartisan Commission on Biodefense Study Panel, ‘Cyberbio convergence: characterizing the multiplicative threat’, convened in Washington DC, 17 Sept. 2019, https://www.youtube.com/watch?v=uQn_xw18Lxc&t=245s. (Unless otherwise noted at point of citation, all URLs cited in this article were accessible on 18 March 2021.) 2 Two sources are particularly useful for understanding the scale of change over the past two decades: National Academies of Sciences, Engineering and Medicine (NASEM), Biodefense in the age of synthetic biology (Washington DC: National Academies Press, 2018), Doi: 10.17226/24890; and Gigi Kwik Gronvall, Synthetic biology: safety, security, and promise (Baltimore, MD: Health Security Press, 2016). 3 For an overview of cyberbiosecurity vulnerabilities, see Jean Peccoud, Jenna E. Gallegos, Randall Murch, Wallace G. Buchholz and Sanjay Raman, ‘Cyberbiosecurity: from naive trust to risk awareness’, Trends in Biotechnology 36: 1, 2018, pp. 4–7, Doi: 10.1016/j.tibtech.2017.10.012. 4 For a full discussion of dual-use research, see NASEM, Governance of dual use research in the life sciences (Washington DC: National Academies Press, 2018), Doi: 10.17226/25154. 5 For a full discussion of strategies and examples of liminal warfare, see David Kilcullen, The dragons and the snakes: how the rest learned to fight the West (Brunswick, Victoria: Scribe, 2020), pp. 115–66. 6 Rory Cormac and Richard J. Aldrich, ‘Grey is the new black: covert action and implausible deniability’, International Affairs 94: 3, 2018, pp. 477–94. 7 Gregory D. Koblentz, ‘Biosecurity reconsidered: calibrating biological threats and responses’, International Security 34: 4, 2010, pp. 96–132, Doi: 10.1162/isec.2010.34.4.96. 8 Defense Advanced Research Projects Agency, Defense-wide justification book, vol. 1 of 5, Research, development, test and evaluation, defense-wide (Washington DC: Department of Defense, 2020), p. 159, https://comptroller.defense.gov/Portals/45/Documents/defbudget/fy2021/budget_justification/pdfs/03_RDT_and_E/RDTE_Vol1_DARPA_MasterJustificationBook_PB_2021.pdf. 9 Hugh Douglas Goold, Philip Wright and Deborah Hailstones, ‘Emerging opportunities for synthetic biology in agriculture’, Genes 9: 7, 2018, p. 341, https://doi.org/10.3390/genes9070341. 10 Kyle B. Justus, Tess Hellebrekers, Daniel D. Lewis et al., ‘A biosensing soft robot: autonomous parsing of chemical signals through integrated organic and inorganic interfaces’, Science Robotics 4: 31, 2019, Doi: 10.1126/scirobotics.aax0765. 11 None of the issues covered in NASEM, Biodefense in the age of synthetic biology, touch on information communication between biological and digital systems. 12 Peccoud et al., ‘Cyberbiosecurity’. 13 Randall S. Murch, William K. So, Wallace G. Buchholz, Sanjay Raman and Jean Peccoud, ‘Cyberbiosecurity: an emerging new discipline to help safeguard the bioeconomy’, Frontiers in Bioengineering and Biotechnology, 5 April 2018, Doi: 10.3389/fbioe.2018.00039. 14 US Bipartisan Commission on Biodefence Study Panel, ‘Cyberbio convergence’. 15 Bruce Schneier, ‘Securing medical research: a cybersecurity point of view’, Science 336: 6088, 2012, pp. 1527–9, Doi: 10.1126/science.1224321. 16 Goold et al., ‘Emerging opportunities for synthetic biology in agriculture’, p. 341. 17 See Zoe Schofield, Gabriel N. Meloni, Peter Tran et al., ‘Bioelectrical understanding and engineering of cell biology’, Journal of the Royal Society, Interface 17: 166, 2020, Doi: 10.1098/rsif.2020.0013. 18 Sarah Vitak, ‘Technology alliance boosts efforts to store data in DNA’, Nature, 3 March 2021, https://www.nature.com/articles/d41586-021-00534-w. 19 Peccoud et al., ‘Cyberbiosecurity’. 20 For a good example, see Justus et al., ‘A biosensing soft robot’. 21 See Johnjoe McFadden and Jim Al-Khalili, ‘The origins of quantum biology’, Proceedings of the Royal Society 474: 2220, 2018, Doi: 10.1098/rspa.2018.0674. On sonics, see Sina Faezi, Sujit Rokka Chhetri, Arnav Vaibhav Malawade et al., ‘Oligo-snoop: a non-invasive side channel attack against DNA synthesis machines’, paper presented to Network and Distributed Systems Security (NDSS) Symposium, 24–7 Feb. 2019, Doi: 10.14722/ndss.2019.23544. Bioelectrochemistry is a common information communication mechanism in living organisms, and, more specifically, a key communication substrate for mammalian brains and nervous systems. See Schofield et al., ‘Bioelectrical understanding and engineering of cell biology’. 22 Thomas A. Dixon, Thomas C. Williams and Isak S. Pretorius, ‘Sensing the future of bio-informational engineering’, Nature Communications 21: 388, 2021, Doi: 10.1038/s41467-020-20764-2. 23 Sara E. Davies and Clare Wenham, ‘Why the COVID-19 response needs International Relations’, International Affairs 96: 5, 2020, pp. 1227–51, Doi: 10.1093/ia/iiaa135. 24 For instance, computer system control of natural and engineered cellular functions via optogenetic or bioelectrical signalling. 25 For a full discussion of the different aspects of cyberbiosecurity, see Randall S. Murch and Diane DiEuliis, eds, Mapping the cyberbiosecurity enterprise (Lausanne: Frontiers Media, 2019). 26 James J. Wirtz, ‘Nuclear disarmament and the end of the chemical weapons “system of restraint”’, International Affairs 95: 4, 2018, pp. 785–800. 27 See Matthew P. Shearer, Michael Montague, Amanda Kobakovich, Matthew Watson, Elena Martin, Gigi Kwik Gronvall and Nancy Cornell, 2nd Annual Global Forum on Scientific Advances Important to the Biological Weapons Convention (Baltimore, MD: John Hopkins Centre for Health Security Project Team, Sept. 2020), https://www.centerforhealthsecurity.org/our-work/events/2019-global-forum/200925-2019GlobalForumMtgRpt.pdf. 28 Kilcullen, The dragons and the snakes, pp. 115–66; Michael Chertoff, Patrick Bury and Daniela Richterova, ‘Bytes not waves: information communication technologies, global jihadism and counterterrorism’, International Affairs 96: 5, 2020, pp. 1305–25. 29 Colonel Gary P. Corn, ‘Cyber national security: navigating gray-zone challenges in and through cyberspace’, in Christopher M. Ford and Winston S. Williams, eds, Complex battlespaces (Oxford: Oxford University Press, 2019), p. 353. 30 George Popp and Sarah Canna, The characterization and conditions of the gray zone, Office of the Secretary of Defense strategic multilayer assessment report (Washington DC, 2016), http://nsiteam.com/social/wp-content/uploads/2017/01/Final_NSI-ViTTa-Analysis_The-Characterization-and-Conditions-of-the-Gray-Zone.pdf. 31 Mark Hoffman and Martin O. Hofmann, ‘Challenges and opportunities in gray zone “combat”’, in Mark Hoffmann, ed., Advances in cross-cultural decision making, ‘Advances in intelligent systems and computing’, vol. 610 (New York: Springer, 2017), pp. 156–66. 32 Advances in human DNA synthesis are an important precursor technology for this hypothetical scenario, and monitoring dual-use research of concern (DURC) is already embedded in relevant grand challenge research projects. See Jef D. Boeke, George Church, Andrew Hessel et al., ‘The genome project-write’, Science 353: 6295, 2016, pp. 126–7, Doi: 10.1126/science.aaf6850. 33 NASEM, Biodefense in the age of synthetic biology, pp. 69–72. 34 Marc Folcher, Sabine Oesterle, Katharina Zwicky, et al., ‘Mind-controlled transgene expression by a wireless-powered optogenetic designer cell implant’, Nature Communications 5: 1, 2014, pp. 1–11, Doi: 10.1038/ncomms6392. 35 Research that seeks to place chemical, optogenetic or bioelectrical gene expression switches in biological devices may facilitate insights into the biological basis of gene functions; this is known as caging and uncaging genes. Such research should be considered DURC if it involves caging and uncaging pathogenic functions, even in non-pathogenic organisms, given the potential (hypothetical as it may be) of signal-activated bioweapons. 36 Avoiding technological surprise in this area will be essential; however, biological weapons intelligence is notoriously difficult to acquire. See Kathleen M. Vogel, Phantom menace or looming danger? A new framework for assessing bioweapons threats (Baltimore: John Hopkins University Press, 2012). 37 Deborah Scott, Dominic Berry and Jane Calvert, ‘Synthetic biology’, in Sahra Gibbon, Barbara Prainsack, Stephen Hilgartner and Janelle Lamoreaux, eds, Handbook of genomics, health and society (Abingdon: Routledge, 2018), pp. 300–307. 38 Gryphon Scientific and Rhodium Group, China's biotechnology development: the role of US and other foreign engagement (Washington DC: US–China Economic and Security Review Commission, 2019), p. 134. 39 Exemplified by the opening manoeuvres of the 2014 Russian intervention in Ukraine. 40 This could be carried out during the period of time military planners call the ‘shaping’ phase: see Kilcullen, The dragons and the snakes, p. 122. 41 An example of this might be inciting a cytokine storm in a vulnerable (immunocompromised) target: see NASEM, Biodefense in the age of synthetic biology, p. 95. 42 Emile Dirks and James Leibold, Genomic surveillance: inside China's DNA dragnet (Canberra: Australian Strategic Policy Institute, 2020), https://www.aspi.org.au/report/genomic-surveillance; Jinghan Zeng, ‘Artificial intelligence and China's authoritarian governance’, International Affairs 96: 6, 2020, pp. 1441–59. 43 Jane Tiller, ‘If you've given your DNA to a DNA database, US police may now have access to it’, The Conversation, 13 Nov. 2019, https://theconversation.com/if-youve-given-your-dna-to-a-dna-database-us-police-may-now-have-access-to-it-126680. 44 Yaniv Erlich, Tal Shor, Itsik Pe'er and Shai Carmi, ‘Identity inference of genomic data using long-range familial searches’, Science 362: 6415, 2018, p. 690, Doi: 10.1126/science.aau4832; Andrew B. Kennedy and Darren J. Lim, ‘The innovation imperative: technology and US–China rivalry in the twenty-first century’, International Affairs 94: 3, 2018, pp. 553–72. 45 Erlich et al., ‘Identity inference of genomic data using long-range familial searches’, p. 690. 46 Sarah Gerke, Carmel Shachar, Peter R. Chai and I. Glenn Cohen, ‘Regulatory, safety, and privacy concerns of home monitoring technologies during COVID-19’, Nature Medicine 26: 8, 2020, pp. 1176–82. 47 Bonnie C. Wintle, Christian R. Boehm, Catherine Rhodes et al., ‘Point of view: a transatlantic perspective on 20 emerging issues in biological engineering’, eLife, vol. 6, 2017, p. 6, Doi: 10.7554/eLife.30247; Luke Kemp, Laura Adam, Christian R. Boehm, Rainer Breitling et al., ‘Point of view: bioengineering horizon scan 2020’, Elife, vol. 9, 2020, Doi: 10.7554/eLife.54489. 48 US Bipartisan Commission on Biodefense Study Panel, ‘Cyberbio convergence’. 49 See e.g. the UN background information document ‘Review of Developments in the Field of Science and Technology Related to the Convention’, https://www.un.org/disarmament/biological-weapons/science-and-technology. 50 Dixon et al., ‘Sensing the future of bio-informational engineering’. 51 See e.g. Katherine French, ‘Harnessing synthetic biology for sustainable development’, Nature Sustainability 2: 4, 2019, pp. 250–52, Doi: 10.1038/s41893-019-0270-x. 52 Synthetic biology, engineering biology and industrial biotechnology are all disciplinary labels for the process of making biology easier to engineer. See Scott et al., ‘Synthetic biology’. 53 For a full discussion of the historical context to design and solution space exploration in engineering biology, see Thomas A. Dixon and Isak S. Pretorius, ‘Drawing on the past to shape the future of synthetic yeast research’, International Journal of Molecular Sciences 21: 19, 2020, Doi: 10.3390/ijms21197156. 54 Thomas C. Williams, Isak S. Pretorius and Ian T. Paulsen, ‘Synthetic evolution of metabolic productivity using biosensors’, Trends in Biotechnology 34: 5, 2016, pp. 371–81; Thomas C. Williams, Xin Xu, Martin Ostrowski, Isak S. Pretorius and Ian T. Paulsen, ‘Positive-feedback, ratiometric biosensor expression improves high-throughput metabolite-producer screening efficiency in yeast’, Synthetic Biology 2: 1, 2017, Doi: 10.1093/synbio/ysw002. 55 Alexander C. Carpenter, Ian T. Paulsen and Thomas C Williams, ‘Blueprints for biosensors: design, limitations, and applications’, Genes 9: 8, 2018, Doi: 10.3390/genes9080375. 56 Carpenter et al., ‘Blueprints for biosensors’, p. 4. 57 Williams et al., ‘Positive-feedback, ratiometric biosensor expression’. 58 George M. Church and Ed Regis, Regenesis: how synthetic biology will reinvent nature and ourselves (New York: Basic Books, 2014), p. 77. 59 Carpenter et al., ‘Blueprints for biosensors’, pp. 2–3. 60 Justus et al., ‘A biosensing soft robot’. 61 Po-Yu Chen, Albert Yu Min Lin, Yen Shan Lin et al., ‘Structure and mechanical properties of selected biological materials’, Journal of the Mechanical Behavior of Biomedical Materials 1: 3, 2008, pp. 208–26, Doi: 10.1016/j.jmbbm.2008.02.003; John W. C. Dunlop and Peter Fratzl, ‘Multilevel architectures in natural materials’, Scripta Materialia 68: 1, 2013, pp. 8–12. 62 Hashwardhan Poddar, Rainer Breitling and Eriko Takano, ‘Towards engineering and production of artificial spider silk using tools of synthetic biology’, Engineering Biology 4: 1, 2020, p. 1, Doi: 10.1049/enb.2019.0017. 63 Paul Egan, Robert Sinko, Philip R. LeDuc and Sinan Keten, ‘The role of mechanics in biological and bio-inspired systems’, Nature Communications 6: 7418, 2015, Doi: 10.1038/ncomms8418. 64 Faezi et al., ‘Oligo-snoop’. 65 NASEM, Safeguarding the bioeconomy (Washington DC: National Academies Press, 2020), pp. 148–9, Doi: 10.17226/25525. 66 Thom Dixon, Natalie C. Curach and Isak S. Pretorius, ‘Bio-informational futures: the convergence of artificial intelligence and synthetic biology’, EMBO Reports 21: e50036 2020, p. 2, Doi: 10.15252/embr.202050036. 67 Hanchuan Peng, ‘Bioimage informatics: a new area of engineering biology’, Bioinformatics 24: 17, 2008, pp. 1827–36, Doi: 10.1093/bioinformatics/btn346. 68 Gronvall, Synthetic biology, p. 20. 69 National Research Council (NRC), Productivity and cyclicality in semiconductors: trends, implications, and questions: report of a symposium (Washington DC: National Academies Press, 2004). 70 Alison McLennan, Regulation of synthetic biology (Cheltenham: Edward Elgar, 2018), pp. 32–3. 71 See, respectively, NASEM, Safeguarding the bioeconomy, pp. 145–7; presentation by James Diggans, director of Bioinformatics and Biosecurity at Twist Bioscience, US Bipartisan Commission on Biodefense Study Panel, ‘Cyberbio convergence’; James Diggans and Emily Leproust, ‘Next steps for access to safe, secure DNA synthesis’, Frontiers in bioengineering and biotechnology, vol. 7, 2019, p. 86, Doi: 10.3389/fbioe.2019.00086. 72 For more information, see NRC, Seeking security: pathogens, open access, and genome databases (Washington DC: National Academies Press, 2004), Doi: 10.17226/11087. 73 Teselagen is a US commercial example of this: see the website ‘Teselagen Biotechnology’, 2020, https://teselagen.com/. 74 Church and Regis, Regenesis, p. 213. 75 Ryan S. Noyce, Seth Lederman and David H. Evans, ‘Construction of an infectious horsepox virus vaccine from chemically synthesized DNA fragments’, PLoS One 13: 1, 2018, Doi: 10.1371/journal.pone.0188453. 76 Diane DiEuliis, Kavita Berger and Gigi Gronvall, ‘Biosecurity implications for the synthesis of horsepox, an orthopoxvirus’, Health Security 15: 6, 2017, pp. 629–37, Doi: 10.1089/hs.2017.0081. 77 The terms design space and solution space capture the mathematical and biological reality that only a small proportion of all possible metabolic pathways or DNA base code combinations will realize a functional organism. This has particular relevance to developments in artificial intelligence and is partly captured by the disciplinary term xenobiology when it refers to new-to-the-world uses of the biological solution space. 78 Nathan Hillson, Mark Caddick, Yizhi Cai et al., ‘Building a global alliance of biofoundries’, Nature 10: 1, 2019, pp. 1–4, Doi: 10.1038/s41467-019-10079-2. 79 Gordon Moore's original prediction was that the number of transistors on a computer chip would double every 18 months. This prediction was subsequently extrapolated to explain the general trend of exponentially increasing capacity in semiconductors (and other technologies) for an ever decreasing cost. See NRC, Productivity and cyclicality, p. 46. 80 Church and Regis, Regenesis, p. 209. 81 Church and Regis, Regenesis, p. 256. 82 Antonio Regalado, ‘China's BGI says it can sequence a genome for just $100’, MIT Technology Review, 26 Feb. 2020, https://www.technologyreview.com/2020/02/26/905658/china-bgi-100-dollar-genome/. 83 An oligo, or an oligonucleotide, is a short strand of RNA or DNA. Oligos form the building blocks for many engineering biology applications in the same way that transistors form the building blocks of semiconductors. 84 Church and Regis, Regenesis, p. 34. 85 The design–build–test–learn workflow is an essential component of synthetic biology and describes the engineering approach to organism design. See Dixon et al., ‘Bio-informational futures’, p. 2. 86 Nishi Srivastava and Anand Akhila, ‘Biosynthesis of artemisinin—revisited’, Journal of Plant Interactions 6: 4, 2011, pp. 265–73, Doi: 10.1080/17429145.2011.570869. 87 Church and Regis, Regenesis, p. 80. 88 Dixon et al., ‘Bio-informational futures’, p. 1. 89 NRC, Productivity and cyclicality, p. 52. 90 Koblentz, ‘Biosecurity reconsidered’. 91 Warren Chin, ‘Technology, war and the state: past, present and future’, International Affairs 95: 4, 2019, pp. 765–84; Kai Liao, ‘The future war studies community and Chinese revolution in military affairs’, International Affairs 96: 6, 2020, pp. 1327–46; Olivier Schmitt, ‘Wartime paradigms and the future of western military power’, International Affairs 96: 2, 2020, pp. 401–18. Author notes The author thanks Professor Sakkie Pretorius, and Drs Jon Symons, Natalie Curach and Thomas Williams for their insights and comments during the drafting process. The author thanks the journal editors and three anonymous reviewers for their feedback. © The Author(s) 2021. Published by Oxford University Press on behalf of The Royal Institute of International Affairs. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - The grey zone of cyber-biological security
JF - International Affairs
DO - 10.1093/ia/iiab041
DA - 2021-05-10
UR - https://www.deepdyve.com/lp/oxford-university-press/the-grey-zone-of-cyber-biological-security-MX0x7xo0CH
SP - 685
EP - 702
VL - 97
IS - 3
DP - DeepDyve
ER -