Researchers warn AI could pose existential risk to humanity

Researchers warn that advanced artificial intelligence systems could pose existential risks to humanity, with one AI alignment researcher estimating over a 10% chance of human extinction within the next decade. The warnings have intensified calls for stronger safety measures and regulation. Current AI models do not present this level of danger, according to researchers, who distinguish today's systems from future highly autonomous ones. Concerns focus on AI alignment, ensuring systems remain compatible with human values, and the rapid pace of development that has left some industry professionals "genuinely scared." The concept of superintelligence, theoretical systems surpassing human intellectual capabilities, drives major companies like OpenAI and Anthropic, which advocate for safety testing and regulation. However, critics question whether such calls could create barriers for smaller competitors, and no consensus exists on appropriate regulatory models.