With Brent Mittelstadt and Sandra Wachter, Chris coordinates the Governance of Emerging Technologies (GET) Research Programme, investigating the legal, ethical, and social aspects of AI, machine learning, and other emerging technologies.
Dr Chris Russell
Research Associate
Profile
Dr Chris Russell is a Group Leader in Safe and Ethical AI at the Alan Turing Institute, and a Reader in Computer Vision and Machine Learning at the University of Surrey. His recent work on explainability (with Sandra Wachter and Brent Mittelstadt of the OII) is cited in the guidelines to the GDPR and forms part of the TensorFlow “What-if tool”. He was one of the first researchers to propose the use of causal models in reasoning about fairness (namely Counterfactual Fairness), and continues to work extensively on computer vision, where he has won best paper award in two of the largest conferences BMVC and ECCV.
Research Interests:
Algorithmic Fairness, Explainable AI, machine learning, computer vision
Research
Current projects
-
Trustworthiness Auditing for AI
Participants: Dr Brent Mittelstadt, Professor Sandra Wachter, Dr Chris Russell, Dr Netta Winstein
This project will evaluate the effectiveness of accountability tools addressing explainability, bias, and fairness in AI. A ‘trustworthiness auditing meta-toolkit’ will be developed and validated via case studies in healthcare and open science.
-
A Right to Reasonable Inferences in Advertising and Financial Services
Participants: Professor Sandra Wachter, Dr Brent Mittelstadt, Dr Silvia Milano, Dr Johann Laux, Dr Chris Russell
This project uses legal and ethical analysis to establish the requirements for applying a ‘right to reasonable inferences’ in Europe to protect against privacy-invasive and discriminatory automated decision-making in advertising and financial services.
Past projects
-
Explaining black-box decisions
Participants: Professor Sandra Wachter, Dr Brent Mittelstadt, Dr Chris Russell
This project transforms the concept of counterfactual explanations into a practically useful tool for explaining automated black-box decisions.
News
-
AI modelling tool developed by Oxford academics incorporated into Amazon anti-bias software
21 April 2021
A new method to better detect discrimination in AI and machine learning systems created by academics at Oxford Internet Institute, has been implemented by Amazon in their bias toolkit, ‘Amazon SageMaker Clarify’, for use by Amazon Web Services customers.
-
New approach needed to address bias in machine learning and AI systems, say Oxford legal and AI ethics experts
8 March 2021
Leading experts in ethics and law from the Oxford Internet Institute (OII) believe that the ways AI systems are commonly designed to measure bias and fairness clash with the fundamental aims of EU non-discrimination law.
-
New research to explore governance of emerging technologies
16 December 2019
The Oxford Internet Institute, part of the University of Oxford, is undertaking a new research programme exploring the Governance of Emerging Technologies (GET).
Press
-
AI: Decoded: China invests in AI — Algorithmic harms — The gender gap
10 March 2021 Politico
In this edition, we look at how the U.K.’s competition authority is getting ready for the AI age. I also spoke with UNESCO’s Gabriela Ramos about how a global ethical framework for artificial intelligence can help women and other minorities.
-
New Approach Needed to Address Bias in Machine Learning and AI Systems, say Oxford Legal and AI Ethics Experts
9 March 2021 Irish Tech News
New approach needed to address bias in machine learning and AI systems, say Oxford legal and AI ethics experts