There is a risk this study will be leveraged by big tech platforms to exonerate themselves.
- Professor of AI and Behavioral Data Science, University of Technology Sydney
Video essayists on YouTube are trying to beat the manosphere at it’s own game. Here are five insightful channels to watch.
Small language models are like specialised tools in a toolbox, compared to something like ChatGPT that brings the whole workshop.
Research reveals why simple prompts can bypass AI safeguards.
Um estudo de aprendizado de máquina com milhões de publicações online encontrou cinco tipos básicos de usuários, incluindo os “discordantes”, que gostam principalmente de contradizer os outros.
A machine learning study of millions of online posts found five basic kinds of users – including ‘disagreers’ who mainly like to contradict others.
CrowdTangle helped researchers and journalists track viral online falsehoods. Its closure in a key election year is dangerous
New research shows that even after Facebook made changes to stem the tide of dangerous pandemic misinformation, some accounts continued to thrive.
People who dive into misinformation are driven to satisfy an emotional need, according to our new research.
Une recherche démontre les avantages du délai de 24 heures accordé aux plates-formes pour retirer un contenu haineux prévu dans le récent « Digital Services Act » européen.
New EU rules require social media platforms to take down flagged posts within 24 hours – and modelling shows that’s fast enough to have a dramatic effect on the spread of harmful content.
An automatic system to determine political ideology from online posts could be a powerful tool against online radicalisation.
A study of more than 20 million LinkedIn users shows acquaintances you have about 10 friends in common with are the most likely to help you get a new job.
Automation is set to put a lot of people out of work - but machine learning could help them find their next career.
Our research provides the first comprehensive picture of long-term trends in online diversity, drawing on a dataset that’s four times as large as the original Hubble Space Telescope data.
Job ads are 12% lower than expected this year.
After analysing posts from 100,000 Twitter users, our research used big data, machine learning and artificial intelligence to reveal the hidden personality traits underpinning thousands of jobs.
Research shows that hiding the popularity of posts can change what people consume, and even improve the overall quality of content.
Contact Marian-Andrei for
- General
- Media request
- Speaking request
- Consulting / Advising
- Research collaboration
- Research supervision
- Location: Sydney, Australia
- Website
- LinkedIn: marian-andrei-rizoiu-063b19b0
- X (formerly Twitter): @andrei_rizoiu
- Article Feed
- ORCID
- Joined
