David Leslie’s research while affiliated with The Alan Turing Institute and other places

Responding to the rapid roll-out and large-scale commercialization of foundation models, large language models, and generative AI, an emerging body of work is shedding light on the myriad impacts these technologies are having across society. Such research is expansive, ranging from the production of discriminatory, fake and toxic outputs, and privacy and copyright violations, to the unjust extraction of labor and natural resources. The same has not been the case in some of the most prominent AI governance initiatives in the global north like the UK's AI Safety Summit and the G7's Hiroshima process, which have influenced much of the international dialogue around AI governance. Despite the wealth of cautionary tales and evidence of algorithmic harm, there has been an ongoing over-emphasis within the AI governance discourse on technical matters of safety and global catastrophic or existential risks. This narrowed focus has tended to draw attention away from very pressing social and ethical challenges posed by the current brute-force industrialization of AI applications. To address such a visibility gap between real-world consequences and speculative risks, this paper offers a critical framework to account for the social, political, and environmental dimensions of foundation models and generative AI. We identify 14 categories of risks and harms and map them according to their individual, social, and biospheric impacts. We argue that this novel typology offers an integrative perspective to address the most urgent negative impacts of foundation models and their downstream applications. We conclude with recommendations on how this typology could be used to inform technical and normative interventions to advance responsible AI.

AI Accountability in Practice aims to provide resources and training materials to help you and your team establish an end-to-end accountability framework. This will enable you to integrate the ethical values and practical principles, which motivate and steer responsible innovation, into the actual processes that characterise your AI project lifecycle. This workbook is part of the AI Ethics and Governance in Practice series (https://aiethics.turing.ac.uk) co-developed by researchers at The Alan Turing Institute in partnership with key public sector stakeholders.

Project teams frequently engage in tasks pertaining to the technical safety and sustainability of their AI projects. In doing so, they need to ensure that their resultant models are reproducible, robust, interpretable, reliable, performant, and secure. The issue of AI safety is of paramount importance, because possible failures have the potential to produce harmful outcomes and undermine public trust. This work of building safe AI outputs is an ongoing process requiring reflexivity and foresight. To aid teams in this, the workbook introduces the core components of AI Safety (reliability, performance, robustness, and security), and helps teams develop anticipatory and reflective skills which are needed to responsibly apply these in practice. This workbook is part of the AI Ethics and Governance in Practice series (https://aiethics.turing.ac.uk) co-developed by researchers at The Alan Turing Institute in partnership with key public sector stakeholders.

This workbook on Data Stewardship aims to provide resources and training which help you and your team to ethically steward the data you access and utilise by proactively initiating and facilitating responsible data practices. You will learn how to use these tools and how they may be relevant at different stages of the project lifecycle. The tools, approaches, and policies introduced should be discussed with your core team and your stakeholders, and should be clearly documented. Data is essential in developing AI models and systems, forming the core information on which they are trained, and, as such, shaping their knowledge base and epistemic (knowledge-contributing) capacity. For this reason, responsible data stewardship is crucial for developing ethical and responsible AI. This workbook is part of the AI Ethics and Governance in Practice series (https://aiethics.turing.ac.uk) co-developed by researchers at The Alan Turing Institute in partnership with key public sector stakeholders.

The purpose of this workbook is to introduce participants to the principle of AI Explainability. Understanding how, why, and when explanations of AI-supported or -generated outcomes need to be provided, and what impacted people’s expectations are about what these explanations should include, is crucial to fostering responsible and ethical practices within your AI projects. To guide you through this process, we will address essential questions: What do we need to explain? And who do we need to explain this to? This workbook offers practical insights and tools to facilitate your exploration of AI Explainability. By providing actionable approaches, we aim to equip you and your team with the means to identify when and how to employ various types of explanations effectively. This workbook is part of the AI Ethics and Governance in Practice series (https://aiethics.turing.ac.uk) co-developed by researchers at The Alan Turing Institute in partnership with key public sector stakeholders.