Your cart is empty.
This is the more realistic concern today. AI could be used to design biological weapons, automate cyberattacks or warfare, and control autonomous weapons systems. Governments and organizations like the United Nations are already debating limits on "killer robots." In this case, AI isn't the decision-maker—humans are using it dangerously.
This is the scenario people like Nick Bostrom or Elon Musk discuss. The concern: AI becomes far more intelligent than humans and pursues goals that don't align with human survival. A classic thought experiment: an AI told to maximize something harmless (like efficiency) might reshape the world in ways that unintentionally eliminate humans.
Important: This is speculative and depends on future breakthroughs we haven't achieved yet. In theory, a highly advanced AI with control over infrastructure, weapons, or biology could pose an existential risk.