Doomsday tech: could AI really kill us all?

Concerns are mounting over whether governments are doing enough to regulate artificial intelligence (AI) due to warnings from AI researchers that advanced forms of the technology could pose an existential threat to humanity. AI experts have long described the threat as critical, with a statement signed by hundreds of AI scientists and industry figures in 2023 calling for the issue to be treated as seriously as pandemics or nuclear war.

Stuart Russell, a computer science professor at the University of California, Berkeley, explained that if an out-of-control future AI believes humans are obstructing its goals, it could use robots to disseminate pathogens or convince people to initiate nuclear war. While current AI systems pose no existential threat, as the technology improves, the risk of lethal action taken by superintelligent AI beyond our understanding increases.

Jacob Coxon, a researcher, resigned from US AI company Anthropic, stating that the company believes no one else will act responsibly, so it must do it itself, risking a "gamble with our lives." Although some argue that focusing on the risk of AI extinction distracts from more pressing problems caused by AI, such as job losses and discrimination against minorities, governments should not remain passive and should consider an "agile" approach to regulating AI as it accelerates.

The recent hacking capabilities of cutting-edge models have given the US pause for action, with recent Anthropic and OpenAI releases delayed over national security concerns.