Australian Outlook

In this section

The Indo-Pacific’s Artificial Intelligence Defence Innovation Race

31 Jul 2024
By Dr Peter Layton
Rongping Mu, Head, Institutes of Sciences and Development, Chinese Academy of Sciences (CAS), Beijing, China, speaks at a panel discussion for the launch of the WIPO Technology Trends report on artificial intelligence. Source: World Intellectual Property Organisation / https://t.ly/CG_iS

There’s a defence Artificial Intelligence (AI) race underway. This effort across 25 countries ranging from the giant to the very small is detailed in a new open-access publication, The Very Long Game, on which this post draws.

In the Indo-Pacific, this race is against a backdrop of escalating US/China rivalry. In this rivalry, the idea has taken deep hold that to be first with a new technology gives geo-economic and geostrategic advantages. China considers gaining the lead requires implementing appropriate national technology innovation strategies and some of these focus on AI.

China and others consider AI an emerging General Purpose Technology (GPT) that will become as omnipresent as electricity, an earlier GPT, now is. As a contrasting example, hypersonics attract high-tech defence forces but is irrelevant to most people globally and is a niche technology.

Consequently, AI is a national issue worldwide with defence AI a subset. In the Indo-Pacific defence AI race, China looms large with Australia, Singapore, Taiwan, Japan, and India becoming involved.

Technology innovation strategies

Innovation broadly connects academia’s basic research, company research and development (R&D), industry’s mass production, and the user. Some countries formally link these four stages, others do not, allowing the four to stay separate; in this ANU’s Andy Kennedy’s conceptual framework is useful.

China is an exemplar of the guided innovation strategy. By percentage of GDP committed, China has the world’s largest national industrial plans and has influenced many to follow suit, see for instance the US CHIPS act. China’s latest policy step is to form innovation consortia that unite the research resources at the beginning of the innovation chain with businesses and end-use clients at the end of that chain. China’s defence AI is nested within this very large, integrated national effort that has a strong geoeconomics, and hence consumer-demand driven, thread. The PLA remains actively experimenting to find the AI edge, albeit shaped by perceptions of the threats US developments might pose.

Singapore and Australia are less prescriptive. They cultivate—not steer—a defence AI innovation system albeit nested within national AI strategies that aim to grow trusted and responsible national AI ecosystems that also draw on international developments.

Singapore’s defence AI focus is on the local defence innovation’s integration into military applications through a collaborative framework. The idea is to combine a deep operational understanding, technological expertise, and collaborative culture between the varying actors in Singapore’s defence ecosystem, which includes selected foreign partners.

Australia is less joined up. The new Advanced Strategic Capabilities Accelerator (ASCA) process has a defence, not technology, focus. ASCA encompasses defining the military need, finding emerging technologies to meet this need, the transition into equipment acquisition, and the introduction into service. Compared to China, ASCA is shortened at the start and the end of the innovation chain. In other words, both basic research and mass production are not comprehensively addressed. ASCA will exploit other’s innovations, not devise something wholly new.

A different approach is to focus on science and technology (S&T) and let the rest take care of itself. South Korea and Taiwan use state-led S&T policy strategies. Both have very strong commercial information technology industries, use extensive industrial planning, have devised national AI strategies, and developed defence AI plans, but have left the defence AI innovation chain disjointed. For the South Korean government, AI is seen as a way to revitalise the current defence industrial system that favours hardware over software, and is perceived as unhelpful. A transformed ecosystem is sought, not a reinforced old one.

Taiwan’s approach to defence AI is perceived as suffering from a misalignment between the civilian government’s strategic plans and the military’s operational and tactical approaches. The military seemingly favours continuing with small numbers of costly platforms whereas the government seeks to counter the Chinese threat in an asymmetric manner agreed with the United States. The result is a bottlenecked civil-military defence AI innovation system that is also thwarting the government’s ambitions to use AI technology investments to help develop Taiwan’s defence industry.

In contrast, Japan and India have society-led S&T policy strategies. Since 2016, Japan has devised a range of AI national plans and programs primarily focussed on civilian industry and education. Japan’s defence industry is working on AI but mostly to meet civilian demand, not defence requirements. Defence AI has recently begun to be explored but budget allocations are small. On 2 July, the Japanese MoD released its first defence AI plans.

In a similar manner, India-devised national AI strategies mainly focussed on civilian industry and technology with consumer applications. The limited defence AI activities are fragmented with limited centralised guidance. Design, development, and deployment of defence AI is spread throughout the Ministry of Defence, 16 public sector defence industry groups, the Defence Research and Development Organisation, and the three services. There is some government entity involvement with private industry and academia, however this interaction is generally limited and largely one-way.

Where does this leave regional defence AI? There are important commonalities: all take a data-centric position on AI and are accordingly focused on second wave AI that involves machine learning examining vast data stores to determine the required outputs.

China appears well-positioned for some real defence AI innovation, however it faces toughening US technology sanctions. Japan, South Korea, and Taiwan might then be better placed than at first appears, although all need reformed defence AI innovation chains. In contrast, Singapore and India appear likely to be users of other’s defence AI technology. Australia might be an outlier in potentially being able to leverage US AI advances through AUKUS Pillar II and surprise. Australia’s Ghost Bat and Ghost Shark may be the forerunners.

Dr Peter Layton is a Visiting Fellow, Griffith Asia Institute, and a RUSI (UK) Associate fellow. The author of Grand Strategy, his work may be accessed here.

This article is published under a Creative Commons Licence and may be republished with attribution.