【大学院課題例】 自律型部隊（ロボット軍）の調達予算（英語論文サンプル）

下記は、イギリスの大学院で課題用に書いたエッセイです。大学院に通っている方、又は通う予定の方に参考になれば幸いです。（改めて読み返すと修正すべき点が多々あるのですが、参考になるよう、修正せずに掲載します。）

課題

下記のトピックについてSWOT分析をする課題でした。

Increasing the procurement defence budget for autonomous combat units (robot army).

エッセイ

Objective: Risk assessment on increasing the procurement defense budget for autonomous weapon system (AWS)
From Who:  A fellow at a US thinktank
To Who:  Members of the United States Senate Armed Services Subcommittee on Emerging Threats and Capabilities
Significance: On January 1, 2021, The US Congress approved the William M. (Mac) Thornberry National Defense Authorization Act (NDAA) for the Fiscal Year 2021 (FY21), which requested $1.7 billion for autonomy and $800 million for AI.
Implications: FY21 NDAA includes policies that enhance US procurement for AWS, such as enhancing the Joint AI Center (JAIC) responsibility and establishing a steering committee to develop a strategy for the technology investments needed to maintain the technological superiority of the US military.
Outlook: Given the provision of FY21 NDAA that relates to AWS, it is necessary to assess the policy and risks of AWS by considering its strength, weakness, opportunity, and threat.

Sector overview

A difficulty of AWS is that there is no clear internationally agreed definition. The Department of Defense Directive (DODD) 3000.09 defines it as a “weapon system that, once activated, can select and engage targets without further intervention by a human operator.” Nevertheless, DODD 3000.09 also requires “commanders and operators to exercise appropriate levels of human judgment over the use of force.” Therefore, AWS in the paper refers to the human-supervised AWS, which are designed to allow humans to override the operation of the weapon system, not a complete autonomous weapon system. Examples of existing AWS are South Korea’s Samsung SGR-A1, Israel’s Harop, and the UK's Taranis, which conceivably attack targets autonomously after human authorization.

SWOT Analysis on the increase of the procurement budget for AWS

Strengths

AWS has several strengths compared to human soldiers. First, AWS can reduce the risk of casualties by removing US soldiers from dangerous and life-threatening missions. Second, by improving the accuracy, it is expected to reduce civilian casualties and other collateral damage. Third, the cost of the procurement of AWS is inexpensive compared to that of keeping human warfighters on the battlefield. For instance, the annual cost of keeping a soldier in Afghanistan costs $850,000. In contrast, the range of procuring existing AWS is from $200,000 to $400,000. Fourth, AWS, fighting at machine speed, will better predict and defeat adversaries, making it more likely to complete its mission and protect the US homeland and citizens.

Furthermore, AWS is more suitable for “dull, dirty or dangerous missions” than human soldiers. An example of a dull mission is the continuous string of nearly day-long MQ-1 Predator missions over Afghanistan and Iraq, while human crews rotated every four-hour duty cycle. An example of a dirty mission is unmanned B-17s to collect radioactive samples by flying into nuclear clouds right after bomb detonation in 1946. An example of a dangerous mission is neutralizing over 11,100 improvised explosive devices in Iraq. These examples indicate that dull, dirty, and dangerous missions are more appropriate for AWS than human soldiers.

Weakness

Some scientists point out that it is difficult for AWS to distinguish a civilian and a soldier; thus, allowing AWS to make decisions on targeting adversaries will potentially result in civilian casualties and collateral damage. Therefore, AWS is most likely deployed in environments where civilians are not present, such as air-to-air combat or naval combat.

Additionally, while a human mistake is normally contained to an individual level, the failure of AWS may result in far disturbing consequences. AWS would pose a risk of “mass fratricide” if weapons turn against friendly forces due to hacking, enemy manipulation, unexpected interactions with the environment, malfunctions, or software errors. In the most extreme case, AWS could continue to attack inappropriate targets “until it exhausts its magazine.” These risks could be mitigated by reinforcing protection against cyber-attacks and software validation. However, since the system of AWS is extremely complex due to the high degree of code required, the risk of accidents can never be entirely eliminated.

There is also a risk of public opposition against increasing procurement of AWS. According to a public poll by Ipsos, 52% of Americans oppose the use of AWS, compared to only 24% of them who support it. Although some researches indicate that these answers vary based on the scenario and context, without further efforts to nurture the public understanding of AWS, there is a concern for public backlash.

Opportunities

Increasing procurements of AWS in the US military will foster the US high-tech industry across a global market. Concerning public-private partnerships, DoD has initiated its Artificial Intelligence Strategy, including “bold new AI initiatives with larger industrial partners, small start-ups and venture capital firms.” DoD also highlighted its will to make long-term stable funding for academics to research in critical areas and to educate the younger generation of AI talent. These investments in public and academic partnerships will benefit the DoD, US industry, and national competitiveness.

Expanding procurement of AWS will increase military efficiency and decrease the overall military budget. As more AWS is employed, the greater the possibility for AWS to reduce tasks, especially in logistical areas and places where humans are likely at risk. Thereby, the US military can relocate human attention to higher-level reasoning and judgment. Furthermore, although AWS requires human authorization and the need for human remains, the overall budget for the US military can be economized by reducing labor demands in the long term.

Threats

There is a geopolitical risk associated with China and Russia. The US military has invested in autonomy and Artificial Intelligence (AI) in the so-called “Third Offset” strategy. However, other countries, especially China and Russia, have been increasingly investing in AI for military purposes. DoD is concerned that these investments will erode “our technological and operational advantages and destabilize the free and open international order.” Concerning AI, Russian President Vladimir Putin explicitly stated that “whoever becomes the leader in this sphere will become the ruler of the world.” In its top-level AI plan, the Chinese Government also said that “by 2030, we shall make artificial intelligence theory, technology, and application at the world's leading level."

On the other hand, increasing the procurement of AWS may trigger its proliferation among other states and non-state actors. An open letter signed by AI/robotic researchers and CEOs of world-leading companies in 2015 stated: “If any major military power pushes ahead with AI weapon development, a global arms race is virtually inevitable, and the endpoint of this technological trajectory is obvious: autonomous weapons will become the Kalashnikovs of tomorrow.”

Due to AWS’s ethical concerns, approximately 30 countries and 165 nongovernmental organizations called for a preemptive ban on AWS. The international discussion on regulating AWS, including its applicability to international humanitarian law, has been taking place under the Group of Government Experts at the United Nations Convention on Certain Conventional Weapons (UN CCW). While participating in every meeting in 2014-2019, the US government has made clear not to support a ban on AWS because they have military and humanitarian benefits. Russia also has opposed the ban on AWS. China has signaled its support of the ban but is still heavily investing in AWS. The discussion under the UNCCW has deepened the mutual understanding of AWS; however, it is challenging to achieve its aim on such a consensus-based forum. Nevertheless, the outcome of such discussion may hold implications for the US policy on AWS.

Outlook

Despite various weaknesses and threats to increasing the procurement defense budget for AWS, the outlook for AWS remains positive. With regard to its weakness and threats, the strict restriction by human control must remain; however, AWS offers significant strengths and opportunities, such as reducing the risk to the lives of US soldiers and foreign citizens, economizing the long-term defense budget, its superior capability of completing missions and revitalizing the US industry.

It is necessary to mitigate the risks of its weakness and threats and consider its deployment ethically and responsibly. Nonetheless, AWS is critical to maintaining the US technological superiority in the harsh security competition with Russia and China.

Bibliography

Burrows, I. (2018) Made in China 2025: Xi Jinping's plan to turn China into the AI world leader, ABC News, 5 October 2018, [online] Available at: https://www.abc.net.au/news/2018-10-06/china-plans-to-become-ai-world-leader/10332614 [Accessed 20 February 2020].

Clapper, J. (2007). Unmanned Systems Roadmap: 2007-2032, Department of Defense, [online] Available at: https://www.globalsecurity.org/intell/library/reports/2007/dod-unmanned-systems-roadmap_2007-2032.pdf [Accessed 18 February 2020].

Deeney, C. (2019). Six in Ten (61%) Respondents Across 26 Countries Oppose the Use of Lethal Autonomous Weapons Systems, Ipsos, 22 January, 2019, [online] Available at: https://www.ipsos.com/en-us/news-polls/human-rights-watch-six-in-ten-oppose-autonomous-weapons [Accessed 19 February 2020].

Department of Defense Directive 3000.09, Autonomy in Weapon Systems, 8 May 207, [online] Available at: https://www.esd.whs.mil/portals/54/documents/dd/issuances/dodd/300009p.pdf [Accessed 19 February 2020].

Etzioni, A. & Etzioni, O. (2017). Pros and Cons of Autonomous Weapons Systems, Military Review, May-June 2017, pp. 72-81.

Francis, D. (2013). How a New Army of Robots Can Cut the Defense Budget, Fiscal Times, 2 April, 2013, [online] Available at: https://www.thefiscaltimes.com/Articles/2013/04/02/How-a-New-Army-of-Robots-Can-Cut-the-Defense-Budget [Accessed 20 February 2020].

Future of Life Institute, Autonomous Weapons: An Open Letter from AI & Robotics Researchers, 28 July, 2015, [online] Available at: https://futureoflife.org/open-letter-autonomous-weapons/ [Accessed 18 February 2020].

Horowitz, M. (2016). Public opinion and the politics of the killer robots debate, Research and Politics, January-March 2016: 1–8. [online] Available at: https://journals.sagepub.com/doi/pdf/10.1177/2053168015627183 [Accessed 20 February 2020].

Horowitz, M. (2019). When speed kills: Lethal autonomous weapon systems, deterrence and stability, Journal of Strategic Studies, 42:6, pp. 764-788.

Human Centered Artificial Intelligence (2021). Summary of AI Provisions from the National Defense Authorization Act 2021, Stanford University, [online] Available at: https://hai.stanford.edu/policy/policy-resources/summary-ai-provisions-national-defense-authorization-act-2021 [Accessed 16 February 2020].

Human Rights Watch. (2020). Stopping Killer Robots: Country Positions on Banning Fully Autonomous Weapons and Retaining Human Control, August 10, 2020, [online] Available at: https://www.hrw.org/report/2020/08/10/stopping-killer-robots/country-positions-banning-fully-autonomous-weapons-and [Accessed 17 February 2020].

Leys, N. (2018). Autonomous Weapon Systems and International Crisis, Air University Press, Strategic Studies Quarterly, Vol. 12, No.1 (Spring 2018), pp. 48-73.

Lewis, L. (2017). Insights for the Third Offset: Addressing Challenges of Autonomy and Artificial Intelligence in Military Operations, CNA, September 2017, [online] Available at: https://www.cna.org/cna_files/pdf/DRM-2017-U-016281-Final.pdf [Accessed 20 February 2020].

Office of the Under Secretary of Defense. (2020). Defense Budget Overview: United States Department of Defense Fiscal Year 2021 Budget Request, Department of Defense, [online] Available at: https://comptroller.defense.gov/Portals/45/Documents/defbudget/fy2021/fy2021_Budget_Request_Overview_Book.pdf [Accessed 20 February 2020].

Sayler, K. (2020). Defense Primer: U.S. Policy on Lethal Autonomous Weapon System, Congressional Research Service, [online] Available at: https://fas.org/sgp/crs/natsec/IF11150.pdf [Accessed 18 February 2020].

Scharre, P. (2016). Autonomous Weapons and Operational Risk, Center for a New American Security, [online] Available at: https://s3.amazonaws.com/files.cnas.org/documents/CNAS_Autonomous-weapons-operational-risk.pdf?mtime=20160906080515 [Accessed 16 February 2020].

Marchant, G et al. (2010), International Governance of Autonomous Military Robots, Columbia Science and Technology Law Review 12, [online] Available at: file:///Users/kentashimomura/Desktop/International%20Governance%20of%20AWS.pdf [Accessed 18 February 2020].

Miller, N. (2013). Terminator-like killer robots worry UN, stuff, May 31 2013, [online] Available at: http://www.stuff.co.nz/video-pointer/8740534/Terminator-like-killer-robots-worry-UN [Accessed 18 February 2020].

U.S. Department of Defense. (2018). Summary of the 2018 Department of Defense Artificial Intelligence Strategy: Harnessing AI to Advance Our Security and Prosperity, [online] Available at https://media.defense.gov/2019/Feb/12/2002088963/-1/-1/1/SUMMARY-OF-DOD-AI-STRATEGY.PDF [Accessed 20 February 2020].

Vincent, J. (2017). Putin Says the Nation That Leads in AI “Will Be the Ruler of the World”, The Verge, 4 September 2017, [online] Available at: https://www.theverge.com/2017/9/4/16251226/russia-ai-putin-rule-the-world [Accessed 19 February 2020].