The American Artificial Intelligence Initiative
INTRODUCTION
Artificial Intelligence (AI) has become a new topic on the international agenda, challenging geostrategic relations, serving as a tool for diplomats and negotiators. AI is not a single piece of technology; it is the culmination of a series of technological trends. The application scope of AI is extraordinary, and it can be identified in core technology and application sectors.
Generally, the term “AI” is used when a machine simulates functions that humans associate with other human minds, such as learning and problem solving (Moore, 2017). According to Spiegeleire (2017:27-28), AI, a kind of “non-human intelligence”, is the automation of intelligent behavior that is driven by a general study of intelligent agents (both biological and artificial). More concretely, AI can be described as “the science and engineering of making intelligent machines, especially intelligent computer programs” (McCarthy, 2006).
AI is relevant basically to any intellectual task. Specifically, AI will play a critical role in driving change in military, information, economic superiority, and the nature of security risks that will affect the ability of states to bolster states‘ competitiveness and strengthen national security. At a high level, national and international security is seeing applications of AI in a wide range of forms, including AI-based warfare, which can be described as a war of algorithms (Leyton, 2018). The accelerating science, technology, and engineering (ST&E) revolutions in the AI area have a policy, legal, ethical and strategic implications for national security (Healy, 2015).
This article focuses on the US as an influential and wealthy actor in international society and a global leader in developing and using AI. Promoting peace and security in the world and often taking an active role as a global peacemaker, the US possesses significant power and potential to change the scenarios in the world’s political scene. The US is possessed particular policy approaches and involved in many current international issues and conflicts.
We will analyze the strategic implications of AI from the perspective of national security in the US, and try to answer the following questions: Why does the US need a strategy for AI? How is the US accelerating its leadership in AI?
DISCUSSION
The US is an important strategic partner to countries and try to establish peaceful and cooperative relations [12]. In the area of geopolitical competition, the US government attaches great importance to its technology companies to maintain its world leadership [11]. AI development in the US is estimated at 33% of the total level of AI development in the world and over 600 U.S. Department of Defense programs involve AI technology [22].
The US has proclaimed intelligent machines as vital to the future of their national security. They develop autonomous weapon systems and are in a weapon race based on AI [13]. The overall focus of the new cyber strategy is divided into three categories: protection of networks, systems, and information of the US Department of Defense; protection against cyber-attacks, as well as the development of integrated cyber capabilities for military operations [22]. As Michael S. Rogers (as head of the agency and Cyber Command in the US) said in October 2016, “Artificial Intelligence and machine learning – in my opinion – [are] the foundation for the future of cybersecurity. … It’s not if, it’s only time for me.” [20]
Furthermore, the US government has three goals for national security policy toward AI technology: Preserve the US technological leadership, support peaceful use of the technology, and manage catastrophic risks [17]. The US federal government plays an important role in regulations and standards for AI innovations, including fostering trust in applications of AI, providing regulatory guidance for AI, and developing relevant technical standards [4]. They have recently sponsored several significant studies on the future of AI and its implications for governance and national security. There are three major reasons why the US needs a national AI strategy: 1) to boost economic competitiveness; 2) to support defense capabilities; and 3) to overcome market failures, including the provisioning of public goods, that would otherwise slow AI development and adoption (Allen, 2017). For instance, the most significant and successful pilot project in the Department of Defense is Project Maven. Robert Work, as deputy secretary of the Department of Defense at that time, formed the Algorithmic Warfare Cross-Functional Team launched in April 2017 [20]. This project used Google’s AI software in the military’s first major attempt to utilize AI in warfare. The project aimed to help defense intelligence analysts in automatically tracking, tagging, and spying on targets by analyzing video motion data from tactical aerial drone platforms such as the ScanEagle and mid-altitude platforms. It was deployed to support the fight against ISIS in the Middle East.
Furthermore, in May 2018 in the White House Summit on AI and the AI Select Committee, Trump declared he would allow scientists and technologists to freely develop their AI inventions in the US [2], specifically endorsing several areas: AI for American Innovation, AI for American Industry, AI for American Workers, and AI with American Values [6].
The priorities discussed were funding AI research, removing regulatory barriers to the deployment of AI-powered technologies, training the future American workforce, achieving strategic military advantage, leveraging AI for government services, and working with allies to promote AI R&D [21].
If the US wants to compete on a global level, it must face an arms race from a number of countries with national AI skills. It is recorded that 18 countries have launched their national AI strategies [7]. As the first significant step taken by the US, on 11 February 2019, President Trump signed an Executive Order (EO) 13859 outlining the American AI Initiative [10] as part of the USA’s AI strategy. The American AI Initiative is structured to support AI innovations and maintain AI leadership, and specifically innovations that aim to increase security and improve the quality of life. The EO does not include real funding related to its implementation about how departments and agencies can deliver on the ambitious goals of the strategy. On the other hand, EO emphasizes the important policies needed to support AI development and US leadership. Specifically, the EO aims to safeguard the national security of the US and shape the global evolution of AI. To achieve this goal, important steps are needed in line with the OECD AI principles to support AI development. Those steps need to be supported by initiatives in AI policy in the US:
- Investments in AI research and development in collaboration with industry, academia, partners, and non-federal entities through technological breakthroughs in AI, such as:
- Doubling of non-defense AI Research and Development (AI R&D) funding over two years.
- In 2019, the AI R&D strategic plan was renewed.
- Identify new opportunities through the „Federal Data Strategy“ as a framework for best practices and operational principles.
- Reducing barriers to AI innovation by proposing the US AI Regulatory Principles and Federal involvement in the development of AI technical standards under the National Science and Technology Council (NSTC) [18].
- Get involved in the international area.
The US and OECD created a consensus agreement between governments was formed on the basic principles for reliable AI management:
- To promote and support the US AI research and innovation
- Opening the market for the US AI industries
- Ensure AI development is consistent with the US values that include protection of privacy, protection of civil rights, and protection of civil liberties.
Later, in May 2019, the US started to endorse the following OECD [4] AI principles:
- Principles for responsible stewardship of trustworthy AI, include inclusive growth, sustainable, development, and well-being, human-centered values and fairness, transparency and explainability, robustness, security, and safety, and accountability.
- National policies and international cooperation for trustworthy AI include investing in AI research and development, fostering a digital ecosystem for AI, providing an enabling policy, environment for AI, building human capacity, and preparing for labor transition, and international cooperation.
- Developing the use of AI for government services and missions with the establishment of the AI Center of Excellence with the aim of:
- To improve the provision and efficiency of government services
- Keep up with the US values
- Improve the quality of Human Resources which includes skills programs; internship; education in the areas of science, technology, engineering, and mathematics (STEM). This quality improvement is prioritized in internships and job training programs, with the main target actors being AI researchers.
Furthermore, if we analyze the US AI regulatory principles, it includes three main points:
- Ensuring public involvement:
- In the policymaking process
- Building public trust
- Technical policies and decisions based on scientific evidence and feedback from civil society, nonprofits, industry leaders, and the academic community.
- Assess the costs, risks, and benefits of developing and implementing AI applications.
- Evaluating regulatory and non-regulatory approaches to AI with fairness, non-discrimination, transparency, safety, and security in mind.
The government has an important role to play in managing the impact of AI on society. Deep social impacts will arise, ranging from changes in the way data is collected and used to how resources and opportunities are allocated to different groups in society. Thus, it is necessary to have a standard of governance to manage these changes into positive changes and in accordance with American values. Moreover, AI governance standards in the US come from the private sector, as companies such as Google and Microsoft have established their own ethical frameworks. In addition, a large number of businesses and NGOs formulate best practices by entering into partnerships on AI. For instance, Google recommended government guidance for AI-powered products and services [19]. By contrast, The European Union’s established a new set of AI ethics guidelines by the High-Level Group on Artificial Intelligence [5]. This shows the US government’s need for more collaboration with the private sector.
CONCLUSION
The US approach to AI regulation is aimed at the private sector including innovators and entrepreneurs who aim to reduce regulatory uncertainty that hinders private sector innovation and AI technology development, to combat authoritarian use of AI with global innovation hubs between the US and international partners and shaping AI technology development in a manner consistent with shared values.
As President Trump said in his State of the Union Address [23], AI technology benefits the American people and the US innovation ecosystem, and it will be the envy of the world in the future. Nevertheless, with regard to formulating a national strategy – in the AI initiative, where is the US’s position in the international area? In fact, the AI initiative is divided into two broad groups. The first group includes planning and specific funding for AI. Meanwhile, the second group includes guidance documents that outline the overall direction of the country’s objectives. The American AI Initiative occupies the second group with the scope of making specific planning commitments in AI development. Further, we have to keep in mind that America is not the only player in AI technology. China’s position in AI development is very strong as an eternal competitor to the US. To highlight the fact, China embodies the goals of the “Made in China 2025” program with a national AI strategy that is its main strength compared to the US. China’s domestic AI sector’s innovation capacity is supported by substantial financial resources for R&D [7]. Beyond not explaining the implementation of AI, the EO also did not explain the importance of collaborating with international allies and partners. The US ignores and does not take the lead in international negotiations with regards to AI governance. The US does not yet have bilateral agreements with other countries to advance and promote AI.
Some policy positions of the US, resulting in the trade war with China, or the fact that a significant part of the contributions to US’s leadership in AI comes from a predominantly immigrant workforce, are the weak points of the American national strategy as outlined in the EO, and this is very likely to become a threat to the advancement of AI development in the US.
Written by Laurencia Krismadewi
REFERENCES
[1] Accelerating America’s Leadership in Artificial Intelligence. 2019. Available online at: https://www.whitehouse.gov/articles/accelerating-americas-leadership-in-artificial-intelligence/
[2] AI Policy – United States. 2019. Available online at: https://futureoflife.org/ai-policy-united-states/
[3] Allen, Greg, and Taniel Chan. 2017. Artificial Intelligence and National Security. Harvard Kennedy School: United States.
[4] American AI Initiative . Available online at: https://www.whitehouse.gov/wp-content/uploads/2020/02/American-AI-Initiative-One-Year-Annual-Report.pdf p.15 – 22.
[5] Artificial Intelligence Strategies. Available online at: https://medium.com/politics-ai/an-overview-of-national-ai-strategies-2a70ec6edfd
[6] Artificial Intelligence for the American People. Available online at: https://www.whitehouse.gov/ai/
[7] Bieri, Matthias. 2018. Artificial Intelligence: China’s High-Tech Ambitions. CSS Analyses in Security Policy. ETH Zurich
[8] Building an AI World: Report on National and Regional AI Strategies. 2018. Available online at: https://cifar.ca/cifarnews/2018/12/06/building-an-ai-world-report-on-national-and-regional-ai-strategies/
[9] De Spiegeleire, Stephan, Matthijs Maas, and Tim Sweijs. 2017. Artificial Intelligence and the Future of Defense: Strategic Implications. The Hague Centre for Strategic Studies (HCSS): The Netherlands. Available online at: https://hcss.nl/sites/default/files/files/reports/Artificial%20Intelligence%20and%20the%20Future%20of%20Defense.pdf (Accessed on 20 November 2020).
[10] Executive Order on Maintaining American Leadership in Artificial Intelligence. 2019. Available online at: https://www.whitehouse.gov/presidential-actions/executive-order-maintaining-american-leadership-artificial-intelligence/
[11] Forbes. (2020). Why the Race for AI Dominance Is More Global Than You Think. Avalaible online at: https://www.forbes.com/sites/cognitiveworld/2020/02/09/why-the-race-for-ai-dominance-is-more-global-than-you-think/#51a15c05121f (Accessed 18 April 2020).
[12] Foreign Policy and National Security. p. 183. Available online at: http://www.jstor.org/stable/2538582
[13] Healy, Gene. 2015. National Security States. Reason; Research Library. Los Angeles Vol. 46, Iss. 10. p.62-66.
[14] Leyton, P. (2018). Algorithmic Warfare: Applying Artificial Intelligence to Warfighting, Canberra, ACT: Air Power Development Centre.
[15] McCarthy, J.; Minsky, M.L.; Rochester, N.; Shannon, C.E. (2006). A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence. AI Mag. 2006, 27, 12.
[16] Moore, A. Carnegie Mellon Dean of Computer Science on the Future of AI. Available online: https://www.forbes.com/sites/peterhigh/2017/10/30/carnegie-mellon-dean-of-computer-science-onthe-future-of-ai/#3a283c652197 (Accessed on 20 April 2020).
[17] National Security Strategy of the United States of America. 2017. Available online: https://www.whitehouse.gov/wp-content/uploads/2017/12/NSS-Final-12-18-2017-0905.pdf (Accessed 29 April 2020).
[18] NSTC – Office of Science and Technology Policy. Available online at: https://www.whitehouse.gov/ostp/nstc/
OECD. Available online at: https://www.oecd.org/going-digital/ai/principles/
[19] Perspectives on Issues in AI Governance. Available online at: https://ai.google/static/documents/perspectives-on-issues-in-ai-governance.pdf
[20] Project Maven brings AI to the fight against ISIS. 2017. Available online at: https://thebulletin.org/2017/12/project-maven-brings-ai-to-the-fight-against-isis/
[21] The National Artificial Intelligence Research and Development Strategic Plan. 2016. Available online at: https://www.nitrd.gov/news/national_ai_rd_strategic_plan.aspx
[22] U.S. Dept of Defense.2018. https://www.defense.gov/home/features/2015/0415_cyber-strategy/ (Accessed 20 November 2020).
[23] Why the U.S. Needs a Strategy for AI. 2019. Available online at: https://www.whitehouse.gov/articles/u-s-needs-strategy-ai/
Picture source: Rowe, A. (2019, Februar 14). What is Trump´s American AI Initiative? Retrieved from: https://tech.co/news/trump-american-ai-initiative-2019-02
1084