The international community unites against the “catastrophic” threats of AI

The first global security summit around artificial intelligence (AI) has managed to USA and China, two rival powers especially opposed in the technological field, have aligned themselves with much of the international community to warn about the potentially “catastrophic” threats of new technology.

Up to 29 governments, including the European Unionmet this Wednesday at a meeting sponsored by the United Kingdom in which they will debate until tomorrow Thursday, together with representatives of the main global technology companies, plans to identify and mitigate the risks of AI.

The most tangible result of the summit is a joint declaration in which the participating countries, which also include Spain, Germany, France, Brazil, India and Canada, among others, commit to promoting international cooperation to avoid worse omens regarding new intelligent systems.

Political leaders perceive immediate dangers such as the capacity of new technology to generate disinformation, but they especially warn about future “existential” risks for humanity as a whole.

The threats range from facilitating computer attacks on a scale never seen before and giving the keys to terrorist groups to manufacture biological weapons, to the possibility of losing control of an autonomous system, whose objectives could begin to diverge from those previously established. planned by humans.

“Given the rapid and uncertain pace of change in AI, and in the context of an acceleration of investments in technology, we affirm that deepening our understanding of these potential risks and actions to mitigate them is especially urgent,” the countries underline. in the joint statement.

The security summit promoted by the United Kingdom will continue in the future. South Korea will host a virtual “mini-summit” in six months, while France will host a new international meeting in a year.

Elon Musk warns of the risk

The summit in Blentchley Park (central England), the enclave where secret Nazi messages were decoded during World War II, was attended by officials from the main technology companies that have started the revolution in the sector.

OpenAI, Google DeepMind, Anthropic, Meta and Microsoft, among others, participate in debates with academics and politicians, many of them behind closed doors, to try to establish a common vision on the challenges that arise.

One of the stars of the summit is the founder of Tesla and owner of the social network Elon Muskwho during the first day of the summit participated in a large part of the debates and warned that AI is one of the main dangers facing humanity.

“For the first time, we are in a situation where there is something that is going to be smarter than the smartest of humans,” said the American tycoon.

“It’s not clear if we can, in fact, control something like that, but I think we should aim to guide it in a direction that is beneficial to humanity,” added Musk, who opted to create an “independent arbiter” to ensure security. in the development of the AI.

Focus on the present risks

Summit organizers want to focus the dialogue on “frontier AI,” models that are still being developed and that could present “existential” threats, but still hypothetical and undefined.

However, some voices at the summit called for focusing on more pressing immediate dangers.

Among them, the vice president of USAKamala Harris, who stressed the need to ensure that the use of current generative artificial intelligence systems does not disproportionately affect the most vulnerable groups.

“When a senior is kicked out of their health plan because of a glitch in an AI algorithm, isn’t that existential for them? “When a woman is threatened by an abusive partner with deepfake photographs, isn’t that existential for her?” Harris said.