How to Fight Financial Crime with AI

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

• David Sutton - Featurespace
• Wednesday 22 November 2023, 14:00-15:00
• Maxwell Centre.

If you have a question about this talk, please contact James Fergusson.

Fraud and scams are now the most common crimes in the UK, with hundreds of thousands of victims each year. As criminals move to adopt AI to attack consumers, banks and financial institutions are in an AI arms race to improve detection and prevention. In this seminar, researchers from the Innovation Lab at Featurespace will discuss open frontiers in AI research that enable new sources of leverage on the problem. Generative AI methods have revealed their potential as a tool to improve consumer protection, however adapting them for use with structured, entity-oriented time series data is not straightforward. Collaborative AI systems, which operate in a federated manner across banks, have also shown a lot of promise for fraud prevention, but open up new challenges of their own for consumer privacy and data protection, which need to be overcome.

Featurespace is a Cambridge-based AI company that specializes in fighting financial crime. After being spun out of the Engineering Department in 2008, it has grown to employ more than 400 people, and has deployed systems to around 100 banks worldwide. The company recently launched TallierLTM, the first pre-trained Large Transaction Model trained across hundreds of millions of consumer accounts.

This talk is part of the Data Intensive Science Seminar Series series.

This talk is included in these lists:

• Cambridge Astronomy Talks
• Cambridge Centre for Data-Driven Discovery (C2D3)
• Cambridge talks
• Chris Davis' list
• Combined External Astrophysics Talks DAMTP
• Cosmology, Astrophysics and General Relativity
• Institute of Astronomy Extra Talks
• Institute of Astronomy Talk Lists
• Interested Talks
• Maxwell Centre
• Titel: TBC
• Trust & Technology Initiative - interesting events
• bld31
• ndk22's list
• ob366-ai4er
• rp587

Note that ex-directory lists are not shown.