The line between civilian and military research has become more permeable in recent years. The increasing sophistication of weapons systems has prompted speculation of a “third revolution” in warfare, in which information technologies allow war to be fought at an unprecedented speed and intensity. While the darkest imaginings of Hollywood films like “Terminator” remain a distant prospect, researchers are increasingly forced to assess their involvement in programs that raise important moral and ethical questions.
Last week, for example, news that the Korea Advanced Institute of Science and Technology (KAIST) would open an artificial intelligence research center in cooperation with Hanwha Systems, a defense company, prompted 50 AI scientists to call for a boycott of work with KAIST because of fears that the research would “accelerate the arms race to develop [autonomous] weapons.” This will, said the signatories in a letter, “permit war to be fought faster and at a scale greater than ever before. … They have the potential to be weapons of terror. … This Pandora’s box will be hard to close if it is opened.”
The president of KAIST, Shin Sung-chul, responded by noting that the university was “significantly aware” of ethical concerns regarding AI, and it had no intention to develop “lethal autonomous weapons systems or killer robots” or “conduct any research activities counter to human dignity including autonomous weapons lacking meaningful human control.” He added that “As an academic institution, we value human rights and ethical standards to a very high degree.” At the same time, however, KAIST will continue cooperation with Hanwha’s defense business unit.
The KAIST boycott followed reports that Google employees had circulated a letter demanding that their company cease its collaboration with the Pentagon on an AI project to analyze drone footage that can be used in the war against Islamic State. “Google should not be in the business of war,” says the letter, signed by over 3,000 employees. While the company insists that the program is focused on “nonoffensive purposes,” the signatories countered that “We cannot outsource the moral responsibility of our technologies to third parties.”
AI has long been a sensitive subject for researchers. In 2015, thousands of them signed an open letter demanding a ban of weapons “beyond meaningful human control.” Twenty-two countries have since called for a pre-emptive ban on autonomous weapons and the United Nations is discussing the issue this week.
This issue has a special resonance in Japan, where the military application of civilian research has extraordinary sensitivity. In recent years, the Japanese government has promoted collaboration between civilian researchers in science and technology and the defense sector to ensure that Japan will remain on the leading edge of defense capabilities. As part of the effort, the Defense Ministry’s Acquisition, Technology and Logistics Agency launched in 2015 the National Security Technology Research Promotion, which provides grants for research that could be applied to defense equipment, and now has a budget that exceeds ¥10 billion.
That program prompted the Science Council of Japan (SCJ) in March 2017 to adopt a statement restating its earlier rejections of military research — made in 1950 and 1967 — and called on other research organizations to develop evaluation procedures to ensure that their work is not linked to the military. A year later, a SCJ survey showed that approximately one-quarter of 183 universities and research organizations have now established such screening systems. The Japanese Society for Artificial Intelligence drew up ethical guidelines in 2017, but they do not debate the issue of military research.
Last month, Kyoto University announced that it would not carry out military research at the school, arguing that such research could lead to threats to tranquility, human happiness and peace. At the same time, however, a standing committee will be established to discuss whether individual cases are appropriate.
Kyoto University’s ambivalence is not unique. Only 30 of 183 research organizations surveyed allow researchers to apply for Defense Ministry grants. Those organizations argue that a blanket refusal to accept such money could deprive them of essential funds and deter young professionals from joining for fear of being cut off from cutting-edge research. Others assert that research is needed to ensure that Japan can defend itself. Neighboring countries are engaged in such efforts and Japan must keep up. Moreover, Japan’s ally, the United States, is pursuing such initiatives and a failure to keep pace will lock this country out of vital defense and security projects.
There are no easy answers to this problem, but Japan must be prepared to make difficult — and painful — choices.