Uploaded November 2026 | Updated November 2026, 1 hour ago
Can you trust AI-generated code to keep your data private, protect against security vulnerabilities, and work exactly as intended? As AI coding assistants and large language models (LLMs) transform software development, ensuring that the code they produce is safe and reliable is becoming increasingly important.
Harvard SEAS postdoctoral researcher Tzu-Han Hsu is exploring how formal methods—mathematical techniques used to verify software behavior—can help make AI-generated code more secure, reliable, and trustworthy.
Building on her research in formal verification, software security, and privacy, Hsu investigates how mathematical guarantees could be integrated into AI-assisted programming. Her work addresses a fundamental challenge: How can we take advantage of AI's speed and creativity while ensuring that the software it generates protects sensitive information and behaves as intended?
By bringing together artificial intelligence, programming languages, and formal verification, this research aims to help build a future where AI-generated software is not only more powerful and efficient, but also provably trustworthy.
Learn more about Tzu-Han Hsu's research: tzuhancs.github.io
_________________
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 #AICoding #SoftwareSecurity #HarvardSEAS
Can you trust AI-generated code to keep your data private, protect against security vulnerabilities, and work exactly as intended? As AI coding assistants and large language models (LLMs) transform software development, ensuring that the code they produce is safe and reliable is becoming increasingly important.
Harvard SEAS postdoctoral researcher Tzu-Han Hsu is exploring how formal methods—mathematical techniques used to verify software behavior—can help make AI-generated code more secure, reliable, and trustworthy.
Building on her research in formal verification, software security, and privacy, Hsu investigates how mathematical guarantees could be integrated into AI-assisted programming. Her work addresses a fundamental challenge: How can we take advantage of AI's speed and creativity while ensuring that the software it generates protects sensitive information and behaves as intended?
By bringing together artificial intelligence, programming languages, and formal verification, this research aims to help build a future where AI-generated software is not only more powerful and efficient, but also provably trustworthy.
Learn more about Tzu-Han Hsu's research: tzuhancs.github.io
_________________
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 #AICoding #SoftwareSecurity #HarvardSEAS










