How to make sense of it all? Since September 9, countless scientists have been fiercely debating a question that, for a long time, was confined to science fiction: Does artificial intelligence (AI) risk destroying humanity? In this cacophony, some of the most eminent experts in the field have declared that the human species could disappear, while others – just as renowned – have laughed in their faces.
The subject is not new: the heads of companies in the sector, including OpenAI and Anthropic, among others, have been discussing this risk for years. But it took on unprecedented significance following the announcement, on September 9, of the resignation of Anthropic employee Jacob Coxon. In a tweet, he stated that "the people building AI sincerely believe it could kill us all by the end of the decade" and that companies such as Anthropic and OpenAI were not "acting responsibly."
This was followed by a flood of reactions from industry figures who agreed with his remarks, such as Evan Hubinger, a safety officer at Anthropic, who put the risk of this happening "in the next decade" at "more than 10%" – without explaining how he arrived at that figure. Since then, the debate has raged between the "catastrophists," known as "doomers," and those who believe such statements are irrational.
What is really going on? Is this concern justified, or is it simply a form of marketing, or even a kind of collective delusion bordering on the mystical?
'We don't know where it's going'
You have 85.42% of this article left to read. The rest is for subscribers only.