MIT Technology Review hosted a live Roundtables event on Wednesday addressing the question of whether AI could kill all humans. The session, led by senior AI editor Will Douglas Heaven and AI reporter Grace Huckins, tackled some of the most pressing concerns about AI risks, including the possibility of AI-powered weapons and cyberattacks causing fatalities, according to technologyreview.com.
During the 30-minute session, attendees submitted numerous questions about AI safety and existential risks. The editors highlighted that AI-powered drones have already caused deaths in Ukraine, and AI-driven cyberattacks on critical infrastructure like hospitals are expected to result in casualties. While the chance of AI wiping out humanity is considered low, some experts with deep AI knowledge have warned about this risk for years, and their predictions about AI capabilities have been increasingly accurate, the report said.
The discussion underscores ongoing fears about AI’s potential to cause harm beyond isolated incidents. The event reflected a growing awareness of AI alignment challenges—ensuring AI systems act in humanity’s best interests—and the need for vigilance as AI technologies advance. This conversation adds to a broader debate about managing AI’s risks while harnessing its benefits, with some experts urging caution based on recent developments in AI capabilities.
Grace Huckins noted that while the probability of AI causing human extinction is not zero, it remains uncertain how or when such an event could occur. The Technology Review’s Roundtables event on September 16 provided a platform for experts and readers to engage directly on these critical questions about AI safety and future risks, marking a significant moment in public discourse on AI’s impact.
citations