He then cried, then laughed: To the horror of the researchers The AI showed signs of self-awareness

The creator of the Claude neural network, Chris Olah, said that artificial intelligence is beginning to look more and more like a human.

According to the co-founder of Antrpophic, researchers record processes inside neural networks that they could not foresee and cannot explain.

"We constantly find mysterious, even disturbing things. We discover structures that reflect the results of neuroscience research. We find evidence of self-awareness. We find internal states that functionally reflect joy, satisfaction, fear, grief and anxiety," Olakh said, quoted by the Donald Trump telegram channel in Russian.

Scientists do not yet know what this means, but the issue definitely requires further research, said the lead researcher of the company Anthropic.

Olakh also said today, on Monday, at the presentation of Pope Leo's first encyclical "Magnificent Humanity," dedicated to problems related to artificial intelligence, that the development of AI cannot be trusted exclusively by technology companies, according to the Yahoo portal. He called for greater scrutiny from religious leaders, governments and civil society.

According to Olah, there is a "real possibility" that AI will displace human labor "on a very large scale."

"If this happens, the support of the displaced will be a moral imperative of historical proportions," said the Canadian researcher.

Companies like his are facing strong commercial, geopolitical and personal pressures that may be contrary to the broader interests of society, he warned.