Description

Since the release of ChatGPT, AI development has accelerated beyond traditional technological trends, outpacing Moore’s Law and reflecting a compounding growth pattern driven by advances in model architecture, data, and computational power. This rapid expansion has created significant infrastructure challenges, particularly in meeting large language models' computational and energy demands. Companies like NVIDIA play a critical role by supplying the hardware that enables large-scale training and deployment. The strong scaling hypothesis, which holds that increasing model size improves performance, has largely proven true, with larger models exhibiting more advanced capabilities. However, generative AI also presents serious challenges, including bias, hallucinations, and the need for transparency and ethical safeguards. As these technologies become more embedded in society, responsible development and oversight are essential to ensure they serve the public good and minimize harm.

Learning Objectives

1. Describe the accelerated growth of AI development since the release of ChatGPT and how it compares to traditional technological trends such as Moore’s Law.
2. Explain the key drivers behind recent advancements in AI, including improvements in model architecture, data availability, and computational power.
3. Identify the infrastructure and energy challenges associated with scaling large language models.

Speaker

Victor Winter, PhD
Professor, Computer Science, University of Nebraska-Omaha

Terms of Use and Privacy Policy

Recording date: July 17, 2025
Date of original release: July 17, 2025
Most recent review/update: July 17, 2025
Termination date: July 16, 2028