Is Washington losing its edge?

Observers have different opinions about the agreement between the US and China regarding the application of artificial intelligence (AI) in the military .

Observers are divided on the agreement between the US and China regarding the application of artificial intelligence (AI) in the military. (Source: Asia Times)

After the meeting between US President Joe Biden and Chinese President Xi Jinping on November 16 (Vietnam time) on the sidelines of the Asia-Pacific Economic Cooperation (APEC) 2023 Summit Week in San Francisco, California (USA), the two countries achieved many important results, including agreeing to resume military communications, cooperate in combating drugs (especially fentanyl) and discuss risks and measures to manage the safety of artificial intelligence (AI).

“We will bring together experts to discuss the risks and safety issues associated with the use of artificial intelligence,” US President Joe Biden said at a press conference following the US-China summit. “When I work with leaders around the world , they all talk about the impact of artificial intelligence. These are concrete steps in the right direction to determine what is necessary, what is dangerous, and what is acceptable.”

The Biden administration recently released its first executive order on artificial intelligence and is pushing hard for global standards on the use of AI in the military. China has also shown signs of being open to discussions, notably banning the use of AI in the command and control (C2) systems of its nuclear arsenal.

Although President Biden and the White House announcement did not explicitly state the connection between AI and nuclear weapons, experts said before the meeting that this was an important topic of discussion between the US and China.

“China is interested in participating in the discussion to establish rules and norms for AI, and we should welcome that,” said Bonnie Glaser, head of the Indo-Pacific program at the German Marshall Fund.

The problem is not just between the US and China.

After SCMP quoted anonymous sources as saying, "US President Joe Biden and Chinese President Xi Jinping are willing to commit to banning the use of AI in autonomous weapons systems, including the use of unmanned aerial vehicles (UAVs) and the control and deployment of nuclear warheads", public opinion has raised hopes for a joint statement between the US and China on this issue.

However, there is no sign that China or the US will accept binding restrictions on their freedom of action in AI.

This is not just a problem for the US and China. Since February 2023, after the US issued its “Policy Statement on the Responsible Use of Artificial Intelligence in the Military,” the country has been working towards building a global consensus on the development and use of military AI, which applies not only to autonomous weapons such as UAVs but also to applications that use algorithms for intelligence analysis or logistics software.

The US aim is to counter the calls by many peace activists and non-aligned countries for a binding ban on “killer robots”, thereby creating an opportunity for the US and its allies to make “responsible” use of AI, a technology that is developing rapidly and can be widely applied in many fields.

Also in February 2023, the Pentagon undertook a major overhaul of its policy on military AI and autonomous systems. Ambassador Bonnie Denise Jenkins, the US Under Secretary of State for Arms Control and International Security, then announced a “Political Statement on the Responsible Use of Artificial Intelligence and Autonomy in Military Affairs” at the Responsible Artificial Intelligence in Military (REAIM) Summit in The Hague in February 2023.

The purpose of this Statement is to lay out an internationally agreed upon US approach to how militaries can responsibly incorporate AI and autonomy into military operations.

Since then, many other countries have voiced their support for the US, including key allies such as Australia, the UK, France, Germany and South Korea, and others such as Hungary, Libya and Turkey. On November 14, Yonhap news agency reported that the US and 45 other countries issued a joint statement highlighting the “responsible” use of AI in the military.

There have been mixed reactions to the meeting between the two leaders, including the US-China agreement on the military application of artificial intelligence. While some see it as necessary, others say Washington is giving up its advantage. Christopher Alexander, director of analysis at the Pioneer Development Group, questioned the need for the agreement, noting that the US would be giving up the strategic advantage it currently holds.

"This is a bad decision. China is behind the US in artificial intelligence technology. Therefore, going through with the deal means the Biden administration is giving up its strategic advantage," said Mr. Alexander.

Commentator Samuel Mangold-Lenett also questioned whether China would honor such a deal, pointing to its lack of compliance with the Paris Climate Agreement. Meanwhile, Phil Siegel, founder of the CAPTRS Center, said such a deal was necessary, although he said major countries like Russia should also be included.

What does Beijing want?

It is no surprise that China has yet to embrace the US approach. “The country’s diplomatic strategy remains focused on competing with and countering US efforts to set future AI governance standards, especially in the military domain,” said Tong Zhao, an expert on AI.

In addition, according to this expert, in managing new military technologies, China often opposes the approval of "responsible" activities, considering this a "political concept that lacks clarity and objectivity".

“Obviously we would like to see the US move towards clear and strong support for the establishment of legal instruments to restrict autonomous weapons systems. We think that political direction and statements are not enough, and neither are most countries,” said Catherine Connolly, a researcher at Stop Killer Robots, an international group of non-governmental organizations that seeks to ban lethal autonomous weapons.

In recent times, the Group of Governmental Experts (GGE) on autonomous weapons has held several discussions in Geneva on related issues, with the aim of proposing the development and application of a set of laws on this type of weapon as previously applied to chemical weapons. However, these efforts have so far been unsuccessful due to the lack of consensus among countries.

So the anti-AI weapons movement has proposed a draft resolution to the United Nations General Assembly in New York. Rather than calling for an immediate ban — which is certain to fail — the Austrian-sponsored resolution simply “requests the Secretary-General of the United Nations to seek the views of Member States.”

As a result, on November 1, 2023, the United Nations General Assembly adopted Resolution L.56, the first resolution on autonomous weapons, which emphasized “the urgent need for the international community to address the challenges and concerns posed by autonomous weapons systems.” Both businesses, academics and non-governmental organizations submitted reports and officially placed the issue on the UN agenda.

Resolution L.56 was adopted with 164 votes in favor, 5 votes against and 8 abstentions. Of these, China abstained.

Researcher Catherine Connolly said that the fact that the US and most countries voted in favor was a positive sign, but unfortunately China abstained.

However, there are some aspects of the resolution that China disagrees with in terms of its characteristics and definitions. In fact, Beijing tends to use a single, narrow definition of “autonomous weapons,” one that only includes systems that, once deployed, “are not subject to human supervision and cannot be stopped.” This allows China to claim to support a ban, while in fact it excludes the vast majority of autonomous systems that many militaries are currently researching and developing.

The UN General Assembly resolution is not binding, but if the US can attract other countries such as the UK, France and possibly the EU into a comprehensive effort, there could be progress in establishing rules in this area, said scholar James Lewis.

So far, international discussion of a non-binding “political declaration” has actually forced Washington to downplay its ambitions by removing a passage related to giving AI the ability to control nuclear weapons.

Source: https://baoquocte.vn/thoa-thuan-my-trung-quoc-ve-ung-dung-ai-trong-quan-su-washington-co-dang-mat-dan-loi-the-250934.html