Uploaded May 2026 | Updated November 2026, 2 days ago
What’s happening inside the “black box” of modern AI systems? Harvard Applied Mathematics and Computer Science concentrator Ben Choi, describes his thesis project, which explores how large language models (LLMs) internally represent emotion, and how those representations compare to patterns found in human psychology and even human brain activity.
Using techniques from geometric machine learning, this research examines the hidden structures inside AI systems and reveals surprising parallels between AI representations, traditional models of human emotion, and neural signal processing in the brain. Ben also reflects on how his coursework at Harvard helped shape the research and how geometric approaches may help researchers better understand increasingly complex AI systems.
_________________
SEE MORE RESEARCH VIDEOS:
youtube.com/playlist?list=PLtHrXQsUhv82_qUFDPoM-OII84iEz0mDA
_________________
SUBSCRIBE TO OUR MONTHLY RESEARCH NEWSLETTER
harvard.us1.list-manage.com/subscribe?u=598d73b470a40f403bba4fb2c&id=318af7d2ea
_________________
VISIT US ON SOCIAL MEDIA
facebook.com/hseas
instagram.com/harvardseas
twitter.com/hseas
linkedin.com/school/1711270/admin/dashboard
_________________
Hashtags:
#HarvardSEAS #Harvardengineering #Harvard #research #ArtificialIntelligence #MachineLearning #LLM #LargeLanguageModels #AIResearch #DeepLearning #Neuroscience #EmotionalAI #AIInterpretability #ComputerScience #AppliedMath #GeometricMachineLearning #DataScience #NeuralNetworks #HarvardSEAS #TechResearch #FutureOfAI #BrainScience #AIEthics #STEMResearch
What’s happening inside the “black box” of modern AI systems? Harvard Applied Mathematics and Computer Science concentrator Ben Choi, describes his thesis project, which explores how large language models (LLMs) internally represent emotion, and how those representations compare to patterns found in human psychology and even human brain activity.
Using techniques from geometric machine learning, this research examines the hidden structures inside AI systems and reveals surprising parallels between AI representations, traditional models of human emotion, and neural signal processing in the brain. Ben also reflects on how his coursework at Harvard helped shape the research and how geometric approaches may help researchers better understand increasingly complex AI systems.
_________________
SEE MORE RESEARCH VIDEOS:
youtube.com/playlist?list=PLtHrXQsUhv82_qUFDPoM-OII84iEz0mDA
_________________
SUBSCRIBE TO OUR MONTHLY RESEARCH NEWSLETTER
harvard.us1.list-manage.com/subscribe?u=598d73b470a40f403bba4fb2c&id=318af7d2ea
_________________
VISIT US ON SOCIAL MEDIA
facebook.com/hseas
instagram.com/harvardseas
twitter.com/hseas
linkedin.com/school/1711270/admin/dashboard
_________________
Hashtags:
#HarvardSEAS #Harvardengineering #Harvard #research #ArtificialIntelligence #MachineLearning #LLM #LargeLanguageModels #AIResearch #DeepLearning #Neuroscience #EmotionalAI #AIInterpretability #ComputerScience #AppliedMath #GeometricMachineLearning #DataScience #NeuralNetworks #HarvardSEAS #TechResearch #FutureOfAI #BrainScience #AIEthics #STEMResearch






