We aim to foster collaboration between technology and ethics to ensure artificial intelligence benefits society. Through curriculum, discussions, and project work, we promote the ethical development of AI in the community.
Our vision is a future where AI is developed responsibly and is aligned with human values to ethically improve lives. We envision a world where AI systems can augment human capabilities and are applied judiciously to further the common good.
AI safety refers to research and precautions taken to ensure AI systems follow human instructions without causing unintended harm. As it becomes more powerful and self-regulating, it is crucial to develop techniques aimed at aligning these systems with human values to prevent potentially catastrophic outcomes.
Some interesting areas of AI safety research include:
AI Safety also considers the long-term risks that arise from superintelligent AI systems that surpass human capabilities. This also includes potential existential threats posed by uncontrolled AI [The Artificial Intelligence Revolution][Squiggle Maximizer]. Researchers are exploring ways to make AI beneficial through techniques such as goal preservation and capability control.