Dr. Roman V. Yampolskiy, supported by Elon Musk, issues a warning about the potential for AI to cause existential harm due to its uncontrollable nature.

Yampolskiy argues that without proof of control over AI, its self-learning and adaptability create endless safety challenges.

He advocates for AI systems to be designed with features for reversibility, limitation, transparency, and comprehensibility in human language to ensure safety.

Concerns are highlighted regarding the increasing autonomy of AI and the risks of it being used in conflicting or harmful ways.

Yampolskiy's forthcoming book presents humanity with a choice between accepting AI as a beneficial but controlling guardian or maintaining freedom and control by rejecting it.

The need to align superintelligent AI with human values is stressed, while acknowledging the complexity and risks of contradictory or harmful directives from such aligned AI.

Elon Musk, who shares concerns about AI risks, has previously funded Yampolskiy's work.

The story ends with a stark reminder that the future of the universe may be at stake as we confront the challenges of unchecked AI.