According to the UN, artificial intelligence is a potential threat to humanity

To combat fake news and the drift of artificial intelligence, the UN secretary general looks to the "nuclear model".

Artificial intelligence is a particular sector:the only one, one might say, in which its own entrepreneurs travel the world denouncing the risks to humanity.Sam Altman, co-founder and CEO of OpenAI, the company that developed ChatGpt, has been doing this for months, speaking with politicians of all nationalities about the "danger" that AI would have for the survival of human civilization.We often even hear him use the word "extinction", confirming the fears - well-founded or not - of his symbolic character.A few days ago, former Google X chief business officer Mo Gawdat even invited humanity to "not have children" precisely because of the systemic uncertainties that these technologies project onto our future.

The apocalyptic scenario of artificial intelligences

But it's easy to talk about armageddon when AI is involved.It's a terrifying scenario, but also fascinating in its own way, reminiscent of a science fiction film (Terminators, for example);but above all it is a distant possibility.Which could happen.Sooner or later.However, there are risks linked to AI that are much closer and more contemporary than the danger of extinction shouted out by Altman:AI is already used today for disinformation and online propaganda campaigns.Just as millions of people have discovered ChatGpt's capabilities in generating short written texts, in fact, political organizations of all kinds are using them to spread fake news and "noise".And they're just getting started.

Last Monday, in a meeting where disinformation was discussed, the Secretary General of the United Nations António Guterres drew the attention of international politics precisely to this side of AI, supporting the creation of a body specifically designed to monitor the development of these technologies, in the mold of the International Atomic Energy Agency (IAEA).The latter, founded in 1957, at the height of the Cold War, was strongly supported by the UN with the aim of promoting the peaceful use of nuclear energy and preventing its use for military purposes.

The parallel between AI and nuclear power was not accidental:according to the Secretary General, in fact, the widespread diffusion of artificial intelligence would represent "an existential risk for humanity similar to that of a nuclear war".In both cases, in fact, these are advanced technologies that can get out of the hands of researchers or be exploited in terrorism or in the war sector.Guterres therefore seems to agree with part of the AI ​​industry itself but believes the emergency is much closer and more realistic than what dark prophets like Altman say.

The open letter from tech scientists

Guterres noted that "the alarm bells about the most recent forms of generative artificial intelligence are deafening", underlining how the most well-founded fears come from "the developers who designed them".The secretary's reference is probably to the open letter signed by various scientists and entrepreneurs in the technology sector, including Elon Musk, Steve Wozniak and Tristan Harris, last March, where they asked to "pause the development" of this technology "for at least six months", in order to find a way to avoid and contain its negative repercussions.The writer and expert on “friendly” artificial intelligence Eliezer Yudkowsky he replied to the letter, specifying that pausing research in the sector was not enough:you just have to stop it.

Altman himself, as mentioned, has often reiterated the possibility of the most pessimistic scenarios, also winning the sympathy of the US Congress last May, when he was invited to a debate on the topic.According to some, his would be a strategy aimed precisely at terrorizing people regulators telling about killer AI and mass extinctions, to divert attention from the looming and probably already ongoing dangers.Such as the spread of fake news, the creation of "synthetic" media generated with AI, i.e. photos, videos and audio in which political figures say and do inappropriate things, for example.They are the so-called deepfake, which risk polluting the wells of political and cultural discussion, creating an environment in which everything could be be fake and nothing is taken seriously.

Well, these are the dangers of artificial intelligence, not the awakening of Skynet (the rebel super computer network from the film Terminators).Guterres seems to have realized this and presented his idea for an international monitoring agency for the sector.It's a shame that "only member states can create it, not the Secretary of the United Nations", as he himself explained.Before thinking about the apocalypse from a science fiction film, it is worth starting from here.