AI researchers put out videos saying superintelligence is ‘exactly as dangerous as it sounds’

A coalition of prominent artificial intelligence researchers, including current and former employees from leading AI companies, has publicly expressed grave concerns about superintelligent AI systems and their potential existential risks to humanity. In a series of video interviews, these experts characterize superintelligence as "exactly as dangerous as it sounds," marking a significant moment of internal industry dissent regarding AI safety priorities.

Geoffrey Irving, a former researcher at both OpenAI and Google DeepMind, stated in a recent interview that he estimates approximately a 50% chance of human extinction from advanced AI systems. This assessment reflects broader concerns among the interviewed researchers, who represent diverse perspectives within the AI research community. The interviews, compiled by Palisade Research, feature voices from OpenAI, Google DeepMind, and Anthropic—three of the most influential organizations developing cutting-edge AI technologies.

These researchers emphasize that superintelligent systems could pose unprecedented risks that differ fundamentally from traditional technological hazards. Their warnings suggest that current safety measures may be inadequate for containing systems with capabilities far exceeding human intelligence.

• Current AI safety protocols may require substantial enhancement before superintelligent systems are deployed
• Internal concerns at major AI companies suggest potential misalignment between corporate timelines and safety requirements
• Regulatory bodies may face pressure to implement stricter oversight of advanced AI development
• Investment and talent recruitment in AI safety research could accelerate
• Public trust in AI development may be affected by these internal warnings
• The AI industry faces growing questions about responsible development trajectories

These declarations from respected researchers within the field carry significant weight, as they come from individuals intimately familiar with current AI capabilities and trajectories. Rather than external critics, these are insiders expressing doubts about their own industry's direction. Their public statements suggest that superintelligence risks deserve consideration alongside competitive and commercial pressures in AI development. As AI systems become increasingly capable, the gap between technical advancement and safety measures may widen, making these voices crucial for informing policy decisions and institutional priorities in artificial intelligence.