Harvard Data Science Review Podcast Digital Twins and Virtual Twins: What Are They and What Do They Do for Humans?
24 snips
Dec 23, 2025 Rachel Franklin, executive director at Harvard's Center for Geographic Analysis, and Patrick Johnson, EVP at Dassault Systèmes, dive into the world of digital and virtual twins. They explore how digital twins replicate complex systems, particularly in urban environments, and discuss industrial applications like those used in aerospace. The conversation also touches on ethical concerns around data privacy and surveillance, the need for interdisciplinary collaboration, and the potential of these technologies to improve elderly care and social connections.
AI Snips
Chapters
Transcript
Episode notes
Real-Time Replica Requires Models And Live Data
- A digital twin is a near-real-time replica of a complex system combining models and continuous data feeds.
- Rachel Franklin stresses that live data plus systems modeling and compute make a twin valuable for testing without harming reality.
Virtual Twins Enable 'Possible Worlds'
- Patrick Johnson prefers the term "virtual twin" to emphasize exploring possible worlds beyond exact replicas.
- He frames twins as fused models in constant dialogue with real-time data (V+R twin).
Boeing's 777 Was Modeled Virtually First
- Boeing fully modeled the 777 virtually in 1994 and called it a digital mock-up.
- Patrick Johnson notes that virtual modeling cut errors so the first 777 outperformed previous production quality.