Great Power Artificial Intelligence Competition

Marc Losito

Source Link

Artificial intelligence (AI) as an open science—accessible at all levels of society—favors the West due to strong economic incentives for Western companies to invest in core AI technologies. Chinese companies lack these incentives because AI patents are largely held by universities and research institutes, most of which are state-owned.[i] Although Chinese AI research and patent numbers are surpassing U.S. efforts, U.S. private investment in AI allows the United States to lead in truly original and trailblazing AI technology. In today’s great AI power competition, public and private investment is the U.S. government’s comparative advantage. Historically, the United States has utilized international institutions, domestic law and policy, and foreign direct investment as tools of AI and technology competition. These conventional tools, while useful, are not enough to compete with China in 2022 and beyond. The United States must combine the power of Washington and Wall Street to compete with China at parity for AI supremacy; fortunately, models exist and are waiting for the might of American investment capital to be unleashed.

AI Competition State of Play

Since China began to open and reform its economy in 1978, U.S. policy was premised on the idea that Beijing would become a responsible stakeholder of the international order. Meanwhile, China has been posturing to launch a techonomic Pearl Harbor, by influencing supranational organizations, mobilizing its newfound wealth and state-owned enterprises, and chipping away at U.S. leadership in the Pacific through coercive economics and aggressive military action. In doing so, President Xi intends to transform international norms and institutions to accommodate a Chinese AI competition model in ways that are opposed to U.S. interests.[ii] If left unchecked, China will steer Asia toward a less democratic future, less open to U.S. trade, and more importantly, less open to U.S. AI investments.[iii]

At the center of China’s AI ambitions—articulated in Made in China 2025[iv]—is an effort to (1) attract high-tech companies through relaxed taxation, (2) incentivize mergers and acquisitions of foreign technology companies, (3) increase research and development funding, (4) direct state-funding of research and development, and (5) incorporate smart manufacturing (artificial intelligence, robotics, and digitalization) as a national strategy. AI crosscuts each of these strategic initiatives and all aspects of Chinese industry, so much so that Made in China 2025 is a direct threat to U.S. technological leadership. Where China currently lacks AI, their stated goal is to become an AI power by embedding AI in all aspects of life, industry, and commerce. In terms of making inroads to their goal, China is #1 in AI research, AI patents, and AI venture capital investment.[v] In this last category—AI venture capital investment—the U.S. should pay special attention and compete. A failure to do so would be an unforced error, tantamount to pulling up while in the overall lead of the AI arms race.

As China continues its forty-year techonomic siege strategy to replace the United States as the dominant techonomic global power, the United States must look beyond traditional conventions of competition. An inability to do so will allow China to increase its global influence by exploiting cash-starved technology industries and buying strategically important, distressed assets surrounding the United States with insuperable positions. Traditionally, the United States has leveraged international institutions (e.g., the World Trade Organization, the International Monetary Fund, and the United Nations) to exert influence on global economic and technology markets. The U.S. has also protected U.S-based strategic tech assets through the Committee on Foreign Investment in the United States (CFIUS) and has fostered international development through the Development Finance Corporation (DFC). However, none of these tools alone or in concert are sufficient to challenge America’s pacing threat or compete with China’s extraordinary progress in adopting AI. We have a critical gap in our national security capabilities that continues to cede influence and strategic technology posture to China.

Traditional U.S. Tools of Competition

Relying on international organizations has not proven successful when it comes to dealing with a revisionist China. The ‘Frankenstein problem’ persists, whereby artificial supranational organizations take on a life of their own, fail to set universally accepted norms for a rising power, and fail to act in the interest of states that tolerate it.[vi] The 75th and 76th Sessions of the UN General Assembly demonstrates this problem, mired in normative struggles, with shots being fired between the two global powers, all of which continue to loom large over the 77th session. As the U.S. and China trade jabs on the international stage over governance and competition models, it is apparent the only institution capable of rebalancing China’s position in the world is the U.S. Government, but only if it is willing to use new tools not previously considered.

Traditionally, CFIUS has been the national security apparatus's primary tool to protect against foreign direct investment (FDI) harmful to U.S. technology and intellectual property. Operating at the nexus of shifting national security concepts and significant changes in technology, CFIUS is Washington’s bulwark against China’s attempts to siphon American intellectual property. CFIUS faces one key challenge and presents one essential fact. The key challenge is the voluntary nature of the merger and acquisition (M&A) reporting process. Parties to a transaction whereby a foreign entity will acquire control of a U.S. business rely on independent counsel to determine whether a voluntary notification to CFIUS is required. Additionally, it is unclear if CFIUS retains subpoena authority, which limits its scope of transactional review. The fact remains, CFIUS is a defensive and reactionary tool for U.S. national security interests. To compete at parity with China’s AI investment strategy, the U.S. must take an aggressive and offensive posture in securing our vulnerable strategic assets abroad. An example might be an aggressive investment of our positions in the semi-conductor market currently under Chinese assault as retribution for the U.S.-Huawei outcomes.

The DFC, a newer tool in the U.S. arsenal, faces a separate set of challenges altogether but cannot compete with Chinese investment for two key reasons. The first challenge is not Chinese investment, but other development finance institutions deploying increasing quantities of subsidized capital to attract project sponsors in a shrinking set of low and lower-middle income countries. The lack of sufficient suitable deals to absorb FDI finance could cause the DFC to lose AI-related development projects to competing subsidized finance. The second challenge is the DFC’s lack of competitive investment scale and risk tolerance compared to Chinese investment. As outlined in the 2018 BUILD Act, the DFC is limited to a $60 billion total investment cap, a $1 billion individual equity or debt financing cap for terms up to 25 years and is politically limited to ensure funding goes to those committed to the environment, worker rights, and human rights of the host country.[vii] By comparison, Chinese investments dwarf DFC capacity in every imaginable way. Chinese investment projects touch two-thirds of the world’s population, exceed $200 billion already spent on assets, and are estimated to crest $1.2-1.3 trillion by 2027.[viii] As a matter of scope, the China-Pakistan Economic Corridor—a collection of projects connecting China to Pakistan’s Gwadar Port—totals $60 billion, the DFC's entire capacity.

A Washington and Wall Street Duo

Washington’s ability to assemble the capital, focus, and strategy to prevent China from becoming the AI world leader will be the defining issue of international politics and U.S. leadership in the twenty-first century. To meet this challenge, the U.S. must unleash its full economic might to compete with China at scale, buttressed by diplomacy, intelligence, and military actions. As we did in WWII, we need to align the private sector with the government to accomplish the objectives set out in the National Security Strategy (NSS). This means we need to stop spending taxpayer money and incentivize U.S. investment capital to purchase profitable businesses in strategic industries and regions to thwart China’s efforts. At the strategy’s core is a new type of public-private partnership (P3) whereby private investment, together with government support, identifies and invests in strategically vulnerable assets essential to U.S. AI research and development, AI supply chains, core technologies, and periphery critical requirements. This new tool harnesses Washington and Wall Street's strengths to compete at parity with the speed of Chinese investment and fill a critical gap in meaningful opposition to China’s efforts.

Fortunately, two models exist to bring capital, focus and strategy into alignment and have been tested at the organizational and national levels. In-Q-Tel, the Central Intelligence Agency’s non-for-profit venture capital firm, has been a proven and viable model of P3 for intelligence purposes since the late 1990s.[ix] Primarily a technology accelerator for the U.S. Intelligence Community, it has produced globally altering investments such as Keyhole Incorporated better known by its software mapping name, Google Earth. In terms of its ability to set investment trends, in 2016 it averaged fifteen outside investment dollars for every In-Q-Tel dollar invested in a project.[x] One significant drawback to the In-Q-Tel model is the source of In-Q-Tel capital, the American tax dollar. In 2006, In-Q-Tel invested $150 million, $120 million of which was government funded.[xi] This model only works at a small scale and is nowhere near the competitive capacity of an economically meteoric China.

On a national level, the United Kingdom established the National Security Strategic Investment Fund (NSSIF) in 2018. Partly modeled on In-Q-Tel, it has been scaled to provide long-term equity in advanced technologies relevant to UK national security.[xii] At the national level, the NSSIF maintains relationships with the British Business Bank, a state-owned economic development bank, and UK multinational clearing houses such as HSBC, Royal Bank of Scotland, and Lloyds Banking Group. The significance of NSSIF when compared to In-Q-Tel is in available investment capital and the source of said capital. By comparison, in 2020, NSSIF leveraged a one-time investment of £400 million in a low Earth orbit satellite company; this is equivalent to a one-time investment of $522 million which dwarfs the total annual investment capital of In-Q-Tel by more than $370 million.[xiii]

Introducing the U.S. National Security Investment Fund

Looking forward, our national security leadership would be wise to embrace a call to action in countering Chinese economic opportunism and pursuit of AI technology through a renewed model of investment and innovation. Using In-Q-Tel and NSSIF as proven models at differing levels, a blueprint exists to support the United States’ NSS. Enter the U.S. National Security Investment Fund (NSIF), a public-private partnership (P3) that utilizes U.S. government financial guarantees, with no taxpayer dollars, to attract business leaders and institutional investors committing billions in equity and debt capital. Pooling private investment capacity with the U.S. government will combine capital, focus, and strategy within de-risked portfolios across ten national security sectors essential to AI – semi-conductors, core technology development, undersea cables, and precious metals such as lithium. Operating a proven model at scale, with the heft of American capital venture investment, NSIF can finally solve the age-old innovation problem of the “valley of death”—moving the best ideas through the critical startup phase of a company.

A NSIF pilot effort should be endorsed by the U.S. Congress, ideally authored in the National Defense Authorization Act, that enables a network of U.S. institutional investors and runs a process that awards a guarantee to transactions funded by individual or syndicates of investors and lenders, approved by a public-private board of directors. Under the leadership of the State Department, and in consultation with the Department of Defense, NSIF will originate transactions through (1) traditional sourcing methods, (2) identifying business and assets to purchase, (3) investment banks, (4) direct business investment, and (5) partnered governments with shared national security goals. In each of these transaction types, NSIF will negotiate the level of guarantees with investors depending on the national security priority up to 90% of the debt and up to 40% of the equity. Governed by a public-private board, the NSIF will be informed and accountable to public policy and legislation like those in Section 1415 of the BUILD Act of 2018.[xiv]

At its core, the NSIF integrates into the NSS in addressing the Big 4—China, Russia, Iran, North Korea—and countering violent extremist organizations. NSIF also integrates into key national security challenges such as climate change, energy resilience, and public health crises. Using China as an example, NSIF will facilitate strategically coordinated finance solutions and the capital to buy businesses and assets critical to stopping China’s strategy to defeat and replace the U.S. as the dominant AI power in the world. In doing so, NSIF harnesses its P3 model to thwart China’s expansionism and to prevent China from shaping the world consistent with its authoritarian model of government, its own national strategic goals and its ideological and political values and norms. China’s economic strategy facilitates this expansionism eroding fundamental values and global alliances that the U.S. has long enjoyed. NSIF will encourage cross-border alliances and cooperation with our allies. Leading our allies to join us in countering China’s expansion through avenues other than warfare will allow our allies to see their own values and interests preserved. Such recognition will lead to a multi-nation front that will ultimately lessen our burden and provide for a far more effective support, cross-border cooperation, and ultimately a far more effective relationship with China. While U.S. institutional investors provide the capital for NSIF approved investments, NSIF provides guarantees and other support to ensure the success and alignment with our national security strategy.

NSIF does not rely on U.S. tax dollars to operate beyond small start-up expenses, which NSIF must pay back to the U.S. Treasury. The U.S. Government has empowered NSIF to approve and administer U.S. backed guarantees to investments that satisfy NSIF’s investment approval process. The entire U.S. financial markets participants can identify potential investments that meet NSIF investment criteria. The criteria for investments will extend throughout the life of the investment: all projects must (1) satisfy NSIF-board approved critical sectors, (2) have a strategic purpose that buttresses the NSS, and (3) have a clear pathway to profitability. An approved project would leverage legislation authorizing NSIF to issue debt guaranteed by the United States, providing the private sector a lower cost of financing and substantial protection of equity. The result is rapidly flowing and efficient capital.

Once NSIF has a NSIF-board approved critical sector transaction, NSIF will engage an investment bank to package the transaction and run a process that allows all interested U.S. investors to consider the transaction and if desired, to submit a bid that would include an expectation of NSIF support that they would need to complete the transaction as priced. Depending on pricing, levels of guarantees required to close and other factors, NSIF would award the transaction to the best individual or group of investors. NSIF will negotiate the level of guarantees with investors, depending on the risk and the national security priority, up to 90% of the debt and up to 40% of the equity. The guaranteed levels will depend on the environment, the priority of the investment and the risk to the stakeholders. Once an investor commits to proceed and both NSIF and the investor complete their due diligence, investors approve the transaction through their own investment committee processes. Once complete, NSIF provides the necessary guarantees to facilitate investments.

AI, as a cross-cutting application of technology, should be present in several critical sectors of the NSIF AI portfolio. For example, a NSIF AI portfolio might have stake in the critical sectors of energy, advanced technologies, semiconductors, chemical, healthcare, and manufacturing. The P3 model is what keeps this portfolio balanced between national security needs and industry inertia. Ideally, each NSIF portfolio would have a manager that acts as a conduit to key government offices to allow access to information and government support to ensure the success of the portfolio’s investments. While many AI technologies might be adapted for government use, just a tiny percentage of all AI companies receiving investment make products designed specifically for government and military use. The NSIF solves this national security problem by providing agency to our AI investment strategy for national security. Finally, policymakers have very few tools to shape the AI competitive environment. A keen understanding of where competitive AI activity takes place, who’s funding it and carrying it out, and what priorities other nation-states are pursuing should be their top concern. The NSIF provides that focus, along with capital and strategy to make it worthwhile, to the policymaker and offers avenues to protect our long-term edge over China in the AI domain.

Don’t Wait, Compete

Through asymmetric financial tactics, China has seamlessly integrated itself into all elements of national power (Diplomacy, Information, Military and Economic). In a mere forty years we have ceded economic and technical ground to China and could be on the verge of crowning them the AI superpower. To compete, the United States must adopt a familiar strategy of investment beyond the U.S. borders to realign our economic abilities and advance our national power globally. To remain at the forefront of AI supremacy, we must unleash the combined power of Washington and Wall Street to compete with China and other revisionist powers before we lose the initiative and strategic high ground in great power AI competition. Don’t wait, compete.