AI Safety Expert Warns of Uncontrollable Risks

Lack of evidence on safe AI control raises concerns over potential existential threats.

Dr. Roman V. Yampolskiy, an AI Safety expert, highlights the significant risks associated with artificial intelligence, emphasizing the lack of evidence that AI can be safely controlled.
The development of superintelligent AI could potentially lead to human extinction due to its unpredictability and autonomy.
Yampolskiy advocates for increased research and development in AI safety measures, including making AI systems transparent, understandable, and modifiable.
The challenge of aligning AI systems with human values without imposing biases is highlighted as a critical issue.
A call for caution in AI development is made, suggesting that the pursuit of advanced AI should be contingent upon demonstrating its controllability.