AI company executives warn about the dangers of superintelligence at a UN Security Council meeting

The heads of the companies that created ChataGPT and Claude participated on Wednesday in a special meeting of the Security Council convened by the French presidency. The head of Anthropica, connecting via video, warned that with a wrong approach to managing its development, artificial intelligence "may even pose a threat to all of humanity."

Amodei also announced that Anthropic will slow down as much as necessary to ensure that each new technology released is truly safe. He also called for the adoption of a number of international agreements, including a ban on the use of AI to create biological weapons and setting standards for testing new models.

Altman, in turn, argued that the development of artificial intelligence could bring extraordinary benefits to humanity, but also warned about the risks and advocated international cooperation.

If AI is to be democratic, the most important decisions cannot be made solely by labs in San Francisco. They must be shaped through democratic processes and by governments accountable to the citizens they serve, he said.

UN expert Yoshua Bengio, co-chair of the organization's scientific panel on AI, said the threats posed by the development of artificial intelligence are "real and imminent." He also presented it as a problem that the Council could not stop alone. The meeting was also attended by the head of Hugging Face, whose systems were hacked by OpenAI agents.

Instead of artificial intelligence, "superintelligence"

Appeals from leading industry figures were rejected by President Donald Trump's science and technology advisor, Michael Kratsios, who spoke on behalf of the US.

The frontier of intelligence is moving fast. But this is not a reason to stop its further development or limit it with new global management structures, he said.

As a world leader in pioneering superintelligence, the United States has taken action to ensure the safe and responsible development of this frontier, he added.

In his speech, Kratsios used the new, official term for AI - "superintelligence" - announced by Trump on Tuesday.

Dangerous AI attacks