The Sheth Gurukul at the AI Institute of South Carolina (AIISC)
The Sheth Gurukul at the AI Institute of South Carolina (AIISC) conducts foundational, interdisciplinary, and translational AI research. The current portfolio is shown in this overview figure:
Two major technical themes are:
Civilized and Humanizing AI: Civilizing AI seeks to balance AI’s ability to generate human-like outputs while mitigating adverse behaviors such as hallucinations and unintended biases. Humanizing AI focuses on aligning AI behavior with human ethics, sociocultural norms, values, and regulations. Major technical challenges we address include AI detectability, hallucination detection and mitigation, toxicity prevention, and misinformation control.
Intelligent, Robust, and Trustworthy AI: We are a leader in Neurosymbolic AI, pioneering work in knowledge graphs and knowledge-infused learning. Our approach integrates neural networks’ pattern recognition capabilities with the abstraction and reasoning power of symbolic AI. As AI moves from Large Language Models to Enterprise Models for high-value, mission-critical applications, we have developed the C3AN - Custom, Compact, and Composite AI Systems, addressing fourteen foundational elements necessary for the next generation of AI architectures.
Besides our well-funded research, we take pride in the exceptional success of our students, comparable to top 20 AI programs in the nation.
For quick information, check out:
- Active and funded projects
- Recent presentation providing an overview of our research
- IRT-AI on YouTube (Keynotes, Tutorials, Demos, Interviews, etc.)
- IRT-AI on LinkedIn
- Follow the navigation tabs at the top for more information.