by Hena Ershadi, AILA Student Assistant
AI in the Liberal Arts recently hosted an informative discussion on generative AI that brought together students from various disciplines to explore the field of artificial intelligence. The event featured speakers from Ideas 2 Innovations (i2i), with Dhyey Mavani as the chief guest, joined by Ryan Ji and Liam Davis. Their combined expertise in AI Safety, research, and industry experience provided valuable insights into today’s AI landscape. They emphasized AI’s role as an augmentation tool rather than a replacement for established technologies, explaining how it enhances recommendation engines, operational management, and workflow scalability.
The panel also discussed the technical foundations of generative AI, highlighting the importance of GPUs and advanced computing in powering Large Language Models (LLMs). A notable segment featured Deepseek’s approach using a Hierarchical Mixture of Experts (MoE) paradigm—a model architecture that offers an alternative perspective compared to industry leaders like OpenAI, Anthropic, and Google.
The conversation extended beyond technical aspects to address the broader impact of AI. The panel and participants discussed the business and financial implications of emerging technologies, evolving regulatory frameworks, geopolitical tensions, and ethical considerations accompanying technological advancement.
The emergence of Deepseek prompted meaningful discussion about China’s political climate and technological priorities. Panelists examined how Deepseek’s development reflects China’s strategic focus on AI advancement and technological independence in a competitive global environment.
The panel also addressed essential questions about bias in AI, including issues of representation in training data, research team diversity, and the potential for algorithms to perpetuate systemic inequities. These discussions explored how biases manifest differently across different emerging AI technologies and what that implies for users around the globe.
Throughout these discussions, the panel maintained relevance for Amherst students by addressing the challenges and opportunities they might encounter as they enter the workforce.
This discussion served as an effective platform for learning and critical thinking. It encouraged attendees to question, analyze, and consider the future direction of generative AI, highlighting its potential to innovate and enhance existing digital systems.