Anthropic CEO warns of possible threat to humanity from AI
Artificial intelligence (AI) technologies, if control over them is lost, could pose a threat to humanity as a whole.
This warning was issued by Dario Amodei, chief executive officer of the American technology corporation Anthropic.
"I even believe that AI could pose a threat to humanity as a whole," he said at a meeting of the UN Security Council on artificial intelligence.
"AI is developing very quickly. Four years ago, it could barely write a line of code or solve a math problem for high school students. Now it writes most of Anthropic's code and handles world-famous unsolved mathematical problems. This trajectory [of development] needs to continue for only a very short period of time — a year or two, perhaps less — to reach what I call a country of geniuses in a data center," he added.
Advanced AI models can bring humanity a great deal of benefit, but they also entail certain risks, he noted, singling out two main categories of threats. "First, it is the improper use of AI models, for example, by bioterrorists to create biological weapons. The second risk lies in the loss of control [over AI]," he believes.
Amodei stated that Anthropic is ready to "slow down the pace [of AI development] as much as is required to guarantee the safety of any AI technology being released." However, he emphasized that the fight against these problems is "bigger than [the capabilities of] one company and is industry-wide and global in scale."
