In the world, forces are joining forces in the shadow of fear of the influence of AI on the development of chemical weapons

The global confrontation with the fears of using artificial intelligence to develop weapons of mass destruction will soon reach a new peak, with the announcement of an international conference to be held on the subject in October in Morocco.

The conference that will deal with the development of chemical weapons with the help of artificial intelligence will be hosted by King Mohammed VI with funding from Qatar, China, Germany, Spain and South Korea. Scientists and diplomats from many countries are expected to arrive in Rabat, in the shadow of fears that artificial intelligence may help hostile elements develop new toxic substances. “Artificial intelligence poses a high threat at a time of international tensions,” explained Fernando Arias, head of the Organization for the Prohibition of Chemical Weapons, last Friday in Berlin.

The use of chemical weapons on the battlefield has a longer “heritage” than nuclear weapons, which were first used in Hiroshima and Nagasaki in 1945. Thirty years earlier, the use of chlorine gas resulted in the death of approximately 6,000 French soldiers within ten minutes at Ypres in Belgium during the First World War. Later, remember how in March 1988 Saddam Hussein killed about 5,000 Kurds by dropping chemical weapons on the city of Halabja, and how the Assad regime repeatedly used chemical weapons against civilians in the previous decade – while former President Barack Obama threatens to respond to such repeated attacks – An unfulfilled threat.

The one who, at least for the time being, is absent from that conference is Russia, a country with an extensive legacy of using chemical weapons, either on the battlefield or in settling accounts with regime opponents. According to the US Secretary of State, for example, Russia used chloropicin against Ukrainian forces. Beyond that, Russia is on record for using the chemical agent “Novichok” against Alexei Navalny in 2020. He did not die from that event, but a few months ago he died in the prison where he was imprisoned under unknown circumstances.

Is it only a “slight risk”?

At its core, artificial intelligence is not a negative thing, and it is used, among other things, by drug manufacturers to develop the medical industry. However, the world began to wake up from its slumber in 2022, when senior researchers published a shocking article in the journal Nature Machine Intelligence, according to which it took artificial intelligence less than six hours to invent 40,000 potentially lethal molecules. In fact, all the researchers required was to take the artificial intelligence that is used for those positive benefits, and turn it into a kind of “bad actor”.

The global industry is aware of a tool that has great potential to be used for good as well as for bad, and that you need a deep treatment of the issue. OpenAI, the creators of GPT chat, said in February that they are building an early warning system, which will inform if the artificial intelligence is capable of creating a biological weapon. The company claimed that the artificial intelligence only poses a “slight risk” of aiding such threats, but admitted that it is correct to check in order to understand in depth the danger reflected from it.

“Global instability makes the possibility of using weapons of mass destruction not only theoretical,” explains the head of the Organization for the Prevention of Chemical Weapons. “It is impossible to neutralize the risks posed by AI, which we are at the beginning of realizing its potential. Science and technology improve the effectiveness of AI, with AI being the most worrisome. It adds lethality.”

At the same time, Günter Sauter, who is in charge of foreign trade at the German Foreign Ministry, says that pathogens created through artificial intelligence “may pose a serious challenge” to the use of biological weapons. “The world may have to readjust the line between freedom of scientific research and scientific responsibility. We must examine what measures we can take within the framework of arms control,” he was quoted in the Emirati “National”.

In the United States are held

A country that has been working on these dangers for several months is the USA. At the behest of President Joe Biden, the Secretary of Homeland Defense published a document in April titled “Safe and Responsible Use of Artificial Intelligence in the Federal Government, Critical Infrastructure, and the American Economy.”

“New resources to reduce the risks of AI in the areas of chemical, biological, radiological and nuclear threats, were submitted as part of a report to the president,” the Homeland Security Secretariat said. “The report includes artificial intelligence models that may increase biological and chemical threats to the US.” As part of the preparation of that document, the administration in Washington joined forces with academics.

The Assistant Secretary of Homeland Security for the fight against weapons of mass destruction, Marie Ellen Callahan, defined the “responsible use” of artificial intelligence as promising a lot for the development of science and solving urgent challenges to improve national security. However, she qualified that “artificial intelligence requires that we be prepared to quickly reduce its misuse for the development of chemical and biological threats.”