AI experts warn governments not to ignore extinction risks

Artificial intelligence experts are cautioning governments against overlooking the potentially catastrophic risks posed by advanced AI systems. These superintelligent machines could pose an existential threat to humanity, wiping out civilization or intentionally targeting humans by spreading deadly pathogens or inciting nuclear war.

While current AI technologies are not considered a threat, experts warn that as AI becomes more capable, the risks could escalate beyond our understanding. Stuart Russell, a computer science professor at UC Berkeley, compared the potential threat of superintelligent AI to the hypothetical power of chimpanzees to wipe out humans, emphasizing the vast difference in intelligence.

The stakes are clear, but the race to develop AI is intensifying, with companies like Anthropic and OpenAI prioritizing rapid advancement over safety measures. This has raised concerns about the potential for AI to be misused, such as in terror attacks or arms proliferation. Amid these warnings, experts urge governments to take decisive action, potentially even implementing a licensing regime for AI, similar to regulations in critical industries.