University of Cambridge > > NLIP Seminar Series > Fighting Bad Information with AI

Fighting Bad Information with AI

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Michael Schlichtkrull.

Bad information ruins lives. It leads to bad decisions that promote hate, damage people’s health and hurt democracy. That’s why Full Fact campaigns for better information in public life. A key part of this is challenging specific incorrect or misleading claims.

Our fact checking starts with us monitoring newspapers, broadcast news and social media, and we collect claims being made in public. We verify whether these claims are clearly supported by the evidence, and ask people to correct the record when they get things wrong. We develop new technology to counter misleading claims by increasing the speed, scale and impact of fact checking. We are not trying to replace fact checkers with technology, but rather to empower fact checkers with the best tools. After talking with many fact checkers, we’ve identified three key areas where technology can help:

  • Finding the most important thing to be fact checking each day
  • Knowing when someone repeats something they already know to be false
  • Checking things in as close to real-time as possible

In this talk, I will describe the tools we are building and the technology behind them, from simple keyword matching through information retrieval algorithms and large language models. I’ll also describe the journey we’ve been on through the development process, what we’ve learned along the way and where we’re going.

Dr David Corney joined Full Fact in 2019 as a data scientist specialising in natural language processing. He helps bring AI into Full Fact’s tools to better support fact checkers and campaigners. This includes training large language models; training regular machine learning models; gathering and annotating data; and working with fact checkers, academics and software engineers from round the world. David completed his PhD in machine learning 20 years ago and has spent the intervening time working in academia and for tech startups in London. He has contributed to several tools that analyse news articles, social media and other sources of text, as well as projects in visual neuroscience and botanical imaging.

This talk is part of the NLIP Seminar Series series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2023, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity