-
Rule of Thumb: Explaining Artificial Intelligence Systems using Partial Information
Authors:
Kaivalya Rawal,
Daria Onitiu,
Brent Mittelstadt,
Sandra Wachter,
Chris Russell
Abstract:
Explainable Artificial Intelligence (XAI) seeks to explain how an Artificial Intelligence (AI) system arrived at a particular decision. We propose ''Rule of Thumb'' (RoT) explanations, a new approach to XAI based upon a novel formulation that identifies the most relevant features for predicting the behaviour of an AI system, for a particular datapoint. We show how RoT is well-suited to enable XAI…
▽ More
Explainable Artificial Intelligence (XAI) seeks to explain how an Artificial Intelligence (AI) system arrived at a particular decision. We propose ''Rule of Thumb'' (RoT) explanations, a new approach to XAI based upon a novel formulation that identifies the most relevant features for predicting the behaviour of an AI system, for a particular datapoint. We show how RoT is well-suited to enable XAI in: (a) zero-shot classification using large language models (LLMs), (b) auditing of opaque AI systems without model access, and (c) the use of AI in scientific discovery. Additionally, RoT meets specific requirements from leading AI regulations, provides a familiar interface and visualisations for XAI practitioners, is model-agnostic, and is substantially faster than alternatives.
Code available at: https://github.com/KaiRawal/Rule-of-Thumb-Explaining-Artificial-Intelligence-Systems-using-Partial-Information
△ Less
Submitted 13 August, 2026; v1 submitted 11 August, 2026;
originally announced August 2026.
-
The Fallacy of Sustainable Generative AI: Limitations in EU Environmental Regulation of Data Centres and Paths Forward
Authors:
Daria Onitiu,
Sandra Wachter,
Brent Mittelstadt
Abstract:
In the age of Artificial Intelligence (AI), Large Language Models, Generative AI and larger frontier AI models, data centres create a significant environmental burden on electricity grids and fresh water resources. Requiring data centre operators and Big Tech under the recast Energy Efficiency Directive (recast EED) to quantify, report and disclose the facility-level energy and water impacts seems…
▽ More
In the age of Artificial Intelligence (AI), Large Language Models, Generative AI and larger frontier AI models, data centres create a significant environmental burden on electricity grids and fresh water resources. Requiring data centre operators and Big Tech under the recast Energy Efficiency Directive (recast EED) to quantify, report and disclose the facility-level energy and water impacts seems to be a step into the right direction towards more transparency and accountability. Yet when two recast EED approved benchmarks - the Power Usage Effectiveness (PUE) and Water Usage Effectiveness (WUE) - can be skewed to create a false sense on efficiency gains, current EU policy pushing for sustainable hyperscale data centre expansion appears misplaced. This paper argues that current PUE and WUE reporting frameworks illustrate what we term the "efficiency paradox," according to which positive scores require retrofitting larger AI data centres at the expense of energy supply and people's water access. Countering this efficiency paradox requires a new strategy for data centre operators and the EU Commission to demonstrate the ecological gains of optimising for efficiency through individual reporting and additional policy interventions. We make three policy proposals to show how this strategy can be formalised in practice: (i) measures to reveal and certify efficiency improvements, (ii) documentation of trade-offs in PUE and WUE improvements and, (iii) a monitoring framework of their diminishing returns and countereffects over time. Implementing these measures will ensure that the recast EED common rating scheme is fit-for-purpose, balancing sustainability with AI innovation for local communities.
△ Less
Submitted 13 June, 2026;
originally announced July 2026.
-
The Heterogeneous Safety Impacts of Benign Multilingual Fine-Tuning
Authors:
Will Hawkins,
Kaivalya Rawal,
Jonathan Rystrøm,
Stratis Tsirtsis,
Zihao Fu,
Greta Warren,
Ryan Brown,
Eoin Delaney,
Sandra Wachter,
Brent Mittelstadt,
Chris Russell
Abstract:
Fine-tuning a large language model is a ubiquitous method for enhancing its capability on a specific downstream task. However, prior work has shown that this increase in capability comes with a cost: it can increase a model's tendency to respond to unsafe adversarial prompts, even when fine-tuning with non-adversarial data. We present the first comprehensive empirical study of this phenomenon in m…
▽ More
Fine-tuning a large language model is a ubiquitous method for enhancing its capability on a specific downstream task. However, prior work has shown that this increase in capability comes with a cost: it can increase a model's tendency to respond to unsafe adversarial prompts, even when fine-tuning with non-adversarial data. We present the first comprehensive empirical study of this phenomenon in multilingual settings by fine-tuning Llama-3.2, Qwen3, and Gemma-3 models using benign data translated across nine languages. We find that safety outcomes are highly sensitive to both the choice of fine-tuning language and the evaluation language, with adversarial compliance rates increasing four-fold in some settings. Multilingual safety drift is decoupled from general capability metrics, and occurs heterogeneously across languages and models. Fine-tuning in non-English languages often induces smaller internal representational drifts than English, but these shifts lead models to default to either exaggerated compliance or refusal. As such, assessing fine-tuning impacts solely in English provides inadequate assurance for deployment. To facilitate further research into these cross-lingual safety blind spots, we release the Multilingual-Benign-Tune dataset and the SORRY-Bench-Multilingual evaluation suite.
△ Less
Submitted 27 June, 2026;
originally announced June 2026.
-
AI-Mediated Communication Can Steer Collective Opinion
Authors:
Stratis Tsirtsis,
Kai Rawal,
Chris Russell,
Brent Mittelstadt,
Sandra Wachter
Abstract:
Generative artificial intelligence (AI) is increasingly integrated into the online platforms where humans exchange opinions; large language models (LLMs) now polish users' posts on LinkedIn and provide context for content shared on X. While prior work has shown that AI can express biased opinions and shape individuals' opinions during human-AI interactions, less attention has been paid to its infl…
▽ More
Generative artificial intelligence (AI) is increasingly integrated into the online platforms where humans exchange opinions; large language models (LLMs) now polish users' posts on LinkedIn and provide context for content shared on X. While prior work has shown that AI can express biased opinions and shape individuals' opinions during human-AI interactions, less attention has been paid to its influence on collective opinion formation when mediating human-to-human communication. We address this gap via a combination of empirical and theoretical analyses. We show empirically that LLMs from multiple popular families introduce directional biases when instructed to edit human-written texts on contested topics, for example, nudging texts in favor of gun control and against atheism. Building on this observation, we introduce a mathematical model of opinion dynamics in which an AI system sits between users on a social network, transforming the opinions they express and perceive. By analytically characterizing the equilibrium of this model and performing simulations on real social network data, we show that biases introduced by AI in human-to-human communication can be amplified through the network and shift collective opinion in their direction. In light of these findings, we investigate whether such biases are controllable by online platforms. We audit the "Explain this post" feature on X and find evidence of pro-life bias in Grok's outputs on abortion-related content, which we trace back to specific design choices. We conclude with a discussion of the broader implications of our findings in relation to ongoing legislative efforts in the European Union.
△ Less
Submitted 15 May, 2026;
originally announced May 2026.
-
Evaluating the Ability of Explanations to Disambiguate Models in a Rashomon Set
Authors:
Kaivalya Rawal,
Eoin Delaney,
Zihao Fu,
Sandra Wachter,
Chris Russell
Abstract:
Explainable artificial intelligence (XAI) is concerned with producing explanations indicating the inner workings of models. For a Rashomon set of similarly performing models, explanations provide a way of disambiguating the behavior of individual models, helping select models for deployment. However explanations themselves can vary depending on the explainer used, and need to be evaluated. In the…
▽ More
Explainable artificial intelligence (XAI) is concerned with producing explanations indicating the inner workings of models. For a Rashomon set of similarly performing models, explanations provide a way of disambiguating the behavior of individual models, helping select models for deployment. However explanations themselves can vary depending on the explainer used, and need to be evaluated. In the paper "Evaluating Model Explanations without Ground Truth", we proposed three principles of explanation evaluation and a new method "AXE" to evaluate the quality of feature-importance explanations. We go on to illustrate how evaluation metrics that rely on comparing model explanations against ideal ground truth explanations obscure behavioral differences within a Rashomon set. Explanation evaluation aligned with our proposed principles would highlight these differences instead, helping select models from the Rashomon set. The selection of alternate models from the Rashomon set can maintain identical predictions but mislead explainers into generating false explanations, and mislead evaluation methods into considering the false explanations to be of high quality. AXE, our proposed explanation evaluation method, can detect this adversarial fairwashing of explanations with a 100% success rate. Unlike prior explanation evaluation strategies such as those based on model sensitivity or ground truth comparison, AXE can determine when protected attributes are used to make predictions.
△ Less
Submitted 13 January, 2026;
originally announced January 2026.
-
OxonFair: A Flexible Toolkit for Algorithmic Fairness
Authors:
Eoin Delaney,
Zihao Fu,
Sandra Wachter,
Brent Mittelstadt,
Chris Russell
Abstract:
We present OxonFair, a new open source toolkit for enforcing fairness in binary classification. Compared to existing toolkits: (i) We support NLP and Computer Vision classification as well as standard tabular problems. (ii) We support enforcing fairness on validation data, making us robust to a wide range of overfitting challenges. (iii) Our approach can optimize any measure based on True Positive…
▽ More
We present OxonFair, a new open source toolkit for enforcing fairness in binary classification. Compared to existing toolkits: (i) We support NLP and Computer Vision classification as well as standard tabular problems. (ii) We support enforcing fairness on validation data, making us robust to a wide range of overfitting challenges. (iii) Our approach can optimize any measure based on True Positives, False Positive, False Negatives, and True Negatives. This makes it easily extensible and much more expressive than existing toolkits. It supports all 9 and all 10 of the decision-based group metrics of two popular review articles. (iv) We jointly optimize a performance objective alongside fairness constraints. This minimizes degradation while enforcing fairness, and even improves the performance of inadequately tuned unfair baselines. OxonFair is compatible with standard ML toolkits, including sklearn, Autogluon, and PyTorch and is available at https://github.com/oxfordinternetinstitute/oxonfair
△ Less
Submitted 5 November, 2024; v1 submitted 30 June, 2024;
originally announced July 2024.
-
The Unfairness of Fair Machine Learning: Levelling down and strict egalitarianism by default
Authors:
Brent Mittelstadt,
Sandra Wachter,
Chris Russell
Abstract:
In recent years fairness in machine learning (ML) has emerged as a highly active area of research and development. Most define fairness in simple terms, where fairness means reducing gaps in performance or outcomes between demographic groups while preserving as much of the accuracy of the original system as possible. This oversimplification of equality through fairness measures is troubling. Many…
▽ More
In recent years fairness in machine learning (ML) has emerged as a highly active area of research and development. Most define fairness in simple terms, where fairness means reducing gaps in performance or outcomes between demographic groups while preserving as much of the accuracy of the original system as possible. This oversimplification of equality through fairness measures is troubling. Many current fairness measures suffer from both fairness and performance degradation, or "levelling down," where fairness is achieved by making every group worse off, or by bringing better performing groups down to the level of the worst off. When fairness can only be achieved by making everyone worse off in material or relational terms through injuries of stigma, loss of solidarity, unequal concern, and missed opportunities for substantive equality, something would appear to have gone wrong in translating the vague concept of 'fairness' into practice. This paper examines the causes and prevalence of levelling down across fairML, and explore possible justifications and criticisms based on philosophical and legal theories of equality and distributive justice, as well as equality law jurisprudence. We find that fairML does not currently engage in the type of measurement, reporting, or analysis necessary to justify levelling down in practice. We propose a first step towards substantive equality in fairML: "levelling up" systems by design through enforcement of minimum acceptable harm thresholds, or "minimum rate constraints," as fairness constraints. We likewise propose an alternative harms-based framework to counter the oversimplified egalitarian framing currently dominant in the field and push future discussion more towards substantive equality opportunities and away from strict egalitarianism by default. N.B. Shortened abstract, see paper for full abstract.
△ Less
Submitted 12 March, 2023; v1 submitted 5 February, 2023;
originally announced February 2023.
-
The Theory of Artificial Immutability: Protecting Algorithmic Groups Under Anti-Discrimination Law
Authors:
Sandra Wachter
Abstract:
Artificial Intelligence (AI) is increasingly used to make important decisions about people. While issues of AI bias and proxy discrimination are well explored, less focus has been paid to the harms created by profiling based on groups that do not map to or correlate with legally protected groups such as sex or ethnicity. This raises a question: are existing equality laws able to protect against em…
▽ More
Artificial Intelligence (AI) is increasingly used to make important decisions about people. While issues of AI bias and proxy discrimination are well explored, less focus has been paid to the harms created by profiling based on groups that do not map to or correlate with legally protected groups such as sex or ethnicity. This raises a question: are existing equality laws able to protect against emergent AI-driven inequality? This article examines the legal status of algorithmic groups in North American and European non-discrimination doctrine, law, and jurisprudence and will show that algorithmic groups are not comparable to traditional protected groups. Nonetheless, these new groups are worthy of protection. I propose a new theory of harm - "the theory of artificial immutability" - that aims to bring AI groups within the scope of the law. My theory describes how algorithmic groups act as de facto immutable characteristics in practice that limit people's autonomy and prevent them from achieving important goals.
△ Less
Submitted 2 May, 2022;
originally announced May 2022.
-
Why Fairness Cannot Be Automated: Bridging the Gap Between EU Non-Discrimination Law and AI
Authors:
Sandra Wachter,
Brent Mittelstadt,
Chris Russell
Abstract:
This article identifies a critical incompatibility between European notions of discrimination and existing statistical measures of fairness. First, we review the evidential requirements to bring a claim under EU non-discrimination law. Due to the disparate nature of algorithmic and human discrimination, the EU's current requirements are too contextual, reliant on intuition, and open to judicial in…
▽ More
This article identifies a critical incompatibility between European notions of discrimination and existing statistical measures of fairness. First, we review the evidential requirements to bring a claim under EU non-discrimination law. Due to the disparate nature of algorithmic and human discrimination, the EU's current requirements are too contextual, reliant on intuition, and open to judicial interpretation to be automated. Second, we show how the legal protection offered by non-discrimination law is challenged when AI, not humans, discriminate. Humans discriminate due to negative attitudes (e.g. stereotypes, prejudice) and unintentional biases (e.g. organisational practices or internalised stereotypes) which can act as a signal to victims that discrimination has occurred. Finally, we examine how existing work on fairness in machine learning lines up with procedures for assessing cases under EU non-discrimination law. We propose "conditional demographic disparity" (CDD) as a standard baseline statistical measurement that aligns with the European Court of Justice's "gold standard." Establishing a standard set of statistical evidence for automated discrimination cases can help ensure consistent procedures for assessment, but not judicial interpretation, of cases involving AI and automated systems. Through this proposal for procedural regularity in the identification and assessment of automated discrimination, we clarify how to build considerations of fairness into automated systems as far as possible while still respecting and enabling the contextual approach to judicial interpretation practiced under EU non-discrimination law.
N.B. Abridged abstract
△ Less
Submitted 12 May, 2020;
originally announced May 2020.
-
Explaining Explanations in AI
Authors:
Brent Mittelstadt,
Chris Russell,
Sandra Wachter
Abstract:
Recent work on interpretability in machine learning and AI has focused on the building of simplified models that approximate the true criteria used to make decisions. These models are a useful pedagogical device for teaching trained professionals how to predict what decisions will be made by the complex system, and most importantly how the system might break. However, when considering any such mod…
▽ More
Recent work on interpretability in machine learning and AI has focused on the building of simplified models that approximate the true criteria used to make decisions. These models are a useful pedagogical device for teaching trained professionals how to predict what decisions will be made by the complex system, and most importantly how the system might break. However, when considering any such model it's important to remember Box's maxim that "All models are wrong but some are useful." We focus on the distinction between these models and explanations in philosophy and sociology. These models can be understood as a "do it yourself kit" for explanations, allowing a practitioner to directly answer "what if questions" or generate contrastive explanations without external assistance. Although a valuable ability, giving these models as explanations appears more difficult than necessary, and other forms of explanation may not have the same trade-offs. We contrast the different schools of thought on what makes an explanation, and suggest that machine learning might benefit from viewing the problem more broadly.
△ Less
Submitted 4 November, 2018;
originally announced November 2018.
-
Counterfactual Explanations without Opening the Black Box: Automated Decisions and the GDPR
Authors:
Sandra Wachter,
Brent Mittelstadt,
Chris Russell
Abstract:
There has been much discussion of the right to explanation in the EU General Data Protection Regulation, and its existence, merits, and disadvantages. Implementing a right to explanation that opens the black box of algorithmic decision-making faces major legal and technical barriers. Explaining the functionality of complex algorithmic decision-making systems and their rationale in specific cases i…
▽ More
There has been much discussion of the right to explanation in the EU General Data Protection Regulation, and its existence, merits, and disadvantages. Implementing a right to explanation that opens the black box of algorithmic decision-making faces major legal and technical barriers. Explaining the functionality of complex algorithmic decision-making systems and their rationale in specific cases is a technically challenging problem. Some explanations may offer little meaningful information to data subjects, raising questions around their value. Explanations of automated decisions need not hinge on the general public understanding how algorithmic systems function. Even though such interpretability is of great importance and should be pursued, explanations can, in principle, be offered without opening the black box. Looking at explanations as a means to help a data subject act rather than merely understand, one could gauge the scope and content of explanations according to the specific goal or action they are intended to support. From the perspective of individuals affected by automated decision-making, we propose three aims for explanations: (1) to inform and help the individual understand why a particular decision was reached, (2) to provide grounds to contest the decision if the outcome is undesired, and (3) to understand what would need to change in order to receive a desired result in the future, based on the current decision-making model. We assess how each of these goals finds support in the GDPR. We suggest data controllers should offer a particular type of explanation, unconditional counterfactual explanations, to support these three aims. These counterfactual explanations describe the smallest change to the world that can be made to obtain a desirable outcome, or to arrive at the closest possible world, without needing to explain the internal logic of the system.
△ Less
Submitted 21 March, 2018; v1 submitted 1 November, 2017;
originally announced November 2017.