Kathleen M. Carley
Professor Carnegie Mellon University
- Pittsburgh PA
Kathleen M. Carley's research combines cognitive science and computer science to address complex social and organizational problems.
Biography
Areas of Expertise
Media Appearances
AI-Generated Election Ads Are Flooding Your Feed – Why AI Is Making the Truth the First Casualty of Future Elections
Yahoo! News online
2026-07-27
As Carnegie Mellon's Kathleen Carley advises, when content triggers sudden strong emotion, that reaction may signal you're "being played by the story."
CMU expert on AI: There's a constant battle between disinformation and those thwarting it
TribLive online
2025-10-12
Kathleen Carley, a professor in Carnegie Mellon University’s Software and Societal Systems Department, laid out the good and bad side of AI during a presentation Thursday for the American Association of University Women’s Murrysville chapter.
Musk and X are epicenter of US election misinformation, experts say
Daily Mail online
2024-11-04
Kathleen Carley (School of Computer Science) says Musk’s wide reach on X helps spread false election information to other platforms, like Reddit and Telegram. She explains that X acts as a “conduit,” allowing misleading content to move easily from one social media site to another.
Meta Is Keeping Quiet This Election
The Atlantic online
2024-11-04
Kathleen Carley (School of Computer Science) has conducted studies on “pink slime” news, a form of fake news that mimics local journalism and has insights into how misinformation spreads across social media platforms, especially during elections.
Elon Musk slams AI 'bias' and calls for 'TruthGPT.' Experts question his neutrality.
ABC News online
2023-04-19
Further, the responses from AI conversation tools depend heavily on the text with which the model is trained, Kathleen Carley, another professor of computer science at Carnegie Mellon University. "There's this view that the majority of information that it was trained on is more left-leaning and has certain political biases and certain political agendas built into it," Carley said. "That's where that argument is coming from."
Twitter Bots Are Promising Cheap Guns to Anyone Using the N-Word
Rolling Stone print
2023-03-06
Kathleen Carley spoke to concerns that bot accounts on Twitter, using a crude form of racial profiling, were targeting specific accounts with messages promoting gun sales. “This looks like a coordinated attempt at provoking harm,” she said. “There may be multiple motivations: to sow dissension, to increase polarization, to encourage racial prejudice. All of these are achieved by these messages.”
Bot Hunting Is All About the Vibes
Wired Online online
2022-09-30
"Kathleen Carley is a computer science professor at the Institute for Software Research at Carnegie Mellon University who has helped develop two bot-detection tools: BotHunter and BotBuster. Carley defines a bot as 'an account that is run using completely automated software,' a definition that aligns with Twitter’s own."
The McGowan Fellows Program Begins Second Decade Of Partnering With Leading MBA Schools
Forbes online
2021-05-12
The 2020 (virtual) symposium, entitled “Applying Principled Leadership in Uncertain Times,” was hosted by the Tepper School of Business at Carnegie Mellon University. Featured presenters included Kathleen Carley of Carnegie Mellon, who spoke on disinformation in cyberspace, and Nicole Lindsay of the Mastercard Center for Inclusive Growth, who spoke about sustainable social impact in business. The social impact experience took place in Chicago and was focused on youth experiencing homelessness.
Researchers: Nearly Half Of Accounts Tweeting About Coronavirus Are Likely Bots
NPR online
2020-05-20
"We do know that it looks like it's a propaganda machine, and it definitely matches the Russian and Chinese playbooks, but it would take a tremendous amount of resources to substantiate that," said Kathleen Carley, a professor of computer science at Carnegie Mellon University who is conducting a study into bot-generated coronavirus activity on Twitter that has yet to be published.
Media
Social
Industry Expertise
Accomplishments
National Geospatial-Intelligence Agency Academic Award
2018
GEOINT
Simmel Award
2011
International Network for Social Network Analysis
Education
University of Zurich
H.D.
Business, Economics and Informatics
2019
Harvard University
Ph.D.
Mathematical Sociology
1984
Massachusetts Institute of Technology
S.B.
Economics & Political Science
1978
Affiliations
- ACM
- IEEE
- Academy of Management
- INFORMS [ORSA/TIMS]
- AAAS
- Sigma XI
- International Network for Social Networks Analysis
- American Sociological Association
Links
Event Appearances
Socially Influence Campaigns: The Coordination of Events Using Bots and Misinformation,
NSF Prepare workshop: Social, Behavioral, economic and governance aspects of pandemics Virtual
Online Terrorism and Insider Threat
7th Workshop on Research for Insider Threats Virtual
Orchestrating Change with Disinformation and Influence
IEEE 2021
What Lies Beneath
CCAC2022 Virtual
Influence Campaigns
CIC/CogMI/TPS Joint Conference Virtual
Articles
#WhatIsDemocracy: finding key actors in a Chinese influence campaign
Computational and Mathematical Organization Theory2023
The rapid increase in China’s outward digital presence on western social media platforms highlights China’s priorities for promoting pro-Chinese narratives and stories in recent years. Simultaneously, China has increasingly been accused of launching information operations using bot activity, puppet accounts, and other inauthentic activity to amplify its messaging. This paper provides a comprehensive network analysis characterization of the hashtag influence campaign China promoted against the US-hosted Summit on Democracy in December 2021, in addition to methods to identify different types of actors within this type of influence campaign.
A Weakly Supervised Classifier and Dataset of White Supremacist Language
arXiv:2306.157322023
We present a dataset and classifier for detecting the language of white supremacist extremism, a growing issue in online hate speech. Our weakly supervised classifier is trained on large datasets of text from explicitly white supremacist domains paired with neutral and anti-racist data from similar domains. We demonstrate that this approach improves generalization performance to new domains. Incorporating anti-racist texts as counterexamples to white supremacist language mitigates bias.
Bridging online and offline dynamics of the face mask infodemic
BMC Digital Health2023
Online infodemics have represented a major obstacle to the offline success of public health interventions during the COVID-19 pandemic. Offline contexts have likewise fueled public susceptibility to online infodemics. We combine a large-scale dataset of Twitter conversations about face masks with high-performance machine learning tools to detect low-credibility information, bot activity, and stance toward face masks in online conversations. We match these digital analytics with offline data regarding mask-wearing and COVID-19 cases to investigate the bidirectional online-offline dynamics of the face mask infodemic in the United States.
Role of masks in mitigating viral spread on networks
Physical Review E2023
Masks have remained an important mitigation strategy in the fight against COVID-19 due to their ability to prevent the transmission of respiratory droplets between individuals. In this work, we provide a comprehensive quantitative analysis of the impact of mask-wearing. To this end, we propose a novel agent-based model of viral spread on networks where agents may either wear no mask or wear one of several types of masks with different properties (e.g., cloth or surgical). We derive analytical expressions for three key epidemiological quantities: The probability of emergence, the epidemic threshold, and the expected epidemic size.
Investigating coordinated account creation using burst detection and network analysis
Journal of Big Data2023
Democracies around the world face the threat of manipulation of their electorates via coordinated online influence campaigns. Researchers have responded by developing valuable methods for finding automated accounts and identifying false information, but these valiant efforts often fall into a cat-and-mouse game with perpetrators who constantly change their behavior. This has forced several researchers to go beyond the detection of individual malicious actors by instead identifying the coordinated activity that propels potent information operations.
