arXiv subjects

Atoosa Kasirzadeh

Publications and source records attributed to Atoosa Kasirzadeh.

At least 37 records · Page 2Linked to original sources

arXiv2025

Multi-Agent Risks from Advanced AI

The rapid development of advanced AI agents and the imminent deployment of many instances of these agents will give rise to multi-agent systems of unprecedented complexity. These systems pose novel and under-explored risks. In this report, we provide a structured taxonomy of these risks by identifying three key failure modes (miscoordination, conflict, and collusion) based on agents' incentives, as well as seven key risk factors (information asymmetries, network effects, selection pressures, destabilising dynamics, commitment problems, emergent agency, and multi-agent security) that can underpin them. We highlight several important instances of each risk, as well as promising directions to help mitigate them. By anchoring our analysis in a range of real-world examples and experimental evidence, we illustrate the distinct challenges posed by multi-agent systems and their implications for the safety, governance, and ethics of advanced AI.

cs.MA↗

arXiv2025

AI Safety for Everyone

Recent discussions and research in AI safety have increasingly emphasized the deep connection between AI safety and existential risk from advanced AI systems, suggesting that work on AI safety necessarily entails serious consideration of potential existential threats. However, this framing has three potential drawbacks: it may exclude researchers and practitioners who are committed to AI safety but approach the field from different angles; it could lead the public to mistakenly view AI safety as focused solely on existential scenarios rather than addressing a wide spectrum of safety challenges; and it risks creating resistance to safety measures among those who disagree with predictions of existential AI risks. Through a systematic literature review of primarily peer-reviewed research, we find a vast array of concrete safety work that addresses immediate and practical concerns with current AI systems. This includes crucial areas like adversarial robustness and interpretability, highlighting how AI safety research naturally extends existing technological and systems safety concerns and practices. Our findings suggest the need for an epistemically inclusive and pluralistic conception of AI safety that can accommodate the full range of safety considerations, motivations, and perspectives that currently shape the field.

cs.CY↗

arXiv2025

Two Types of AI Existential Risk: Decisive and Accumulative

The conventional discourse on existential risks (x-risks) from AI typically focuses on abrupt, dire events caused by advanced AI systems, particularly those that might achieve or surpass human-level intelligence. These events have severe consequences that either lead to human extinction or irreversibly cripple human civilization to a point beyond recovery. This discourse, however, often neglects the serious possibility of AI x-risks manifesting incrementally through a series of smaller yet interconnected disruptions, gradually crossing critical thresholds over time. This paper contrasts the conventional "decisive AI x-risk hypothesis" with an "accumulative AI x-risk hypothesis." While the former envisions an overt AI takeover pathway, characterized by scenarios like uncontrollable superintelligence, the latter suggests a different causal pathway to existential catastrophes. This involves a gradual accumulation of critical AI-induced threats such as severe vulnerabilities and systemic erosion of economic and political structures. The accumulative hypothesis suggests a boiling frog scenario where incremental AI risks slowly converge, undermining societal resilience until a triggering event results in irreversible collapse. Through systems analysis, this paper examines the distinct assumptions differentiating these two hypotheses. It is then argued that the accumulative view can reconcile seemingly incompatible perspectives on AI risks. The implications of differentiating between these causal pathways -- the decisive and the accumulative -- for the governance of AI as well as long-term AI safety are discussed.

cs.CY↗

arXiv2024

The Only Way is Ethics: A Guide to Ethical Research with Large Language Models

There is a significant body of work looking at the ethical considerations of large language models (LLMs): critiquing tools to measure performance and harms; proposing toolkits to aid in ideation; discussing the risks to workers; considering legislation around privacy and security etc. As yet there is no work that integrates these resources into a single practical guide that focuses on LLMs; we attempt this ambitious goal. We introduce 'LLM Ethics Whitepaper', which we provide as an open and living resource for NLP practitioners, and those tasked with evaluating the ethical implications of others' work. Our goal is to translate ethics literature into concrete recommendations and provocations for thinking with clear first steps, aimed at computer scientists. 'LLM Ethics Whitepaper' distils a thorough literature review into clear Do's and Don'ts, which we present also in this paper. We likewise identify useful toolkits to support ethical work. We refer the interested reader to the full LLM Ethics Whitepaper, which provides a succinct discussion of ethical considerations at each stage in a project lifecycle, as well as citations for the hundreds of papers from which we drew our recommendations. The present paper can be thought of as a pocket guide to conducting ethical research with LLMs.

cs.CL↗

arXiv2024

A Taxonomy of Systemic Risks from General-Purpose AI

Through a systematic review of academic literature, we propose a taxonomy of systemic risks associated with artificial intelligence (AI), in particular general-purpose AI. Following the EU AI Act's definition, we consider systemic risks as large-scale threats that can affect entire societies or economies. Starting with an initial pool of 1,781 documents, we analyzed 86 selected papers to identify 13 categories of systemic risks and 50 contributing sources. Our findings reveal a complex landscape of potential threats, ranging from environmental harm and structural discrimination to governance failures and loss of control. Key sources of systemic risk emerge from knowledge gaps, challenges in recognizing harm, and the unpredictable trajectory of AI development. The taxonomy provides a snapshot of current academic literature on systemic risks. This paper contributes to AI safety research by providing a structured groundwork for understanding and addressing the potential large-scale negative societal impacts of general-purpose AI. The taxonomy can inform policymakers in risk prioritization and regulatory development.

cs.CY↗

arXiv2024

Ethics Whitepaper: Whitepaper on Ethical Research into Large Language Models

This whitepaper offers an overview of the ethical considerations surrounding research into or with large language models (LLMs). As LLMs become more integrated into widely used applications, their societal impact increases, bringing important ethical questions to the forefront. With a growing body of work examining the ethical development, deployment, and use of LLMs, this whitepaper provides a comprehensive and practical guide to best practices, designed to help those in research and in industry to uphold the highest ethical standards in their work.

cs.CY↗

arXiv2024

Measurement challenges in AI catastrophic risk governance and safety frameworks

Safety frameworks represent a significant development in AI governance: they are the first type of publicly shared catastrophic risk management framework developed by major AI companies and focus specifically on AI scaling decisions. I identify six critical measurement challenges in their implementation and propose three policy recommendations to improve their validity and reliability.

cs.CY↗

arXiv2024

Foundational Challenges in Assuring Alignment and Safety of Large Language Models

This work identifies 18 foundational challenges in assuring the alignment and safety of large language models (LLMs). These challenges are organized into three different categories: scientific understanding of LLMs, development and deployment methods, and sociotechnical challenges. Based on the identified challenges, we pose $200+$ concrete research questions.

cs.LG↗

arXiv2024

Beyond Model Interpretability: Socio-Structural Explanations in Machine Learning

What is it to interpret the outputs of an opaque machine learning model. One approach is to develop interpretable machine learning techniques. These techniques aim to show how machine learning models function by providing either model centric local or global explanations, which can be based on mechanistic interpretations revealing the inner working mechanisms of models or nonmechanistic approximations showing input feature output data relationships. In this paper, we draw on social philosophy to argue that interpreting machine learning outputs in certain normatively salient domains could require appealing to a third type of explanation that we call sociostructural explanation. The relevance of this explanation type is motivated by the fact that machine learning models are not isolated entities but are embedded within and shaped by social structures. Sociostructural explanations aim to illustrate how social structures contribute to and partially explain the outputs of machine learning models. We demonstrate the importance of sociostructural explanations by examining a racially biased healthcare allocation algorithm. Our proposal highlights the need for transparency beyond model interpretability, understanding the outputs of machine learning systems could require a broader analysis that extends beyond the understanding of the machine learning model itself.

cs.LG↗

arXiv2024

The Future of Open Human Feedback

Human feedback on conversations with language language models (LLMs) is central to how these systems learn about the world, improve their capabilities, and are steered toward desirable and safe behaviors. However, this feedback is mostly collected by frontier AI labs and kept behind closed doors. In this work, we bring together interdisciplinary experts to assess the opportunities and challenges to realizing an open ecosystem of human feedback for AI. We first look for successful practices in peer production, open source, and citizen science communities. We then characterize the main challenges for open human feedback. For each, we survey current approaches and offer recommendations. We end by envisioning the components needed to underpin a sustainable and open human feedback ecosystem. In the center of this ecosystem are mutually beneficial feedback loops, between users and specialized models, incentivizing a diverse stakeholders community of model trainers and feedback providers to support a general open feedback pool.

cs.HC↗

arXiv2024

Epistemic Injustice in Generative AI

This paper investigates how generative AI can potentially undermine the integrity of collective knowledge and the processes we rely on to acquire, assess, and trust information, posing a significant threat to our knowledge ecosystem and democratic discourse. Grounded in social and political philosophy, we introduce the concept of \emph{generative algorithmic epistemic injustice}. We identify four key dimensions of this phenomenon: amplified and manipulative testimonial injustice, along with hermeneutical ignorance and access injustice. We illustrate each dimension with real-world examples that reveal how generative AI can produce or amplify misinformation, perpetuate representational harm, and create epistemic inequities, particularly in multilingual contexts. By highlighting these injustices, we aim to inform the development of epistemically just generative AI systems, proposing strategies for resistance, system design principles, and two approaches that leverage generative AI to foster a more equitable information ecosystem, thereby safeguarding democratic values and the integrity of knowledge production.

cs.AI↗

arXiv2024

A Review of Modern Recommender Systems Using Generative Models (Gen-RecSys)

Traditional recommender systems (RS) typically use user-item rating histories as their main data source. However, deep generative models now have the capability to model and sample from complex data distributions, including user-item interactions, text, images, and videos, enabling novel recommendation tasks. This comprehensive, multidisciplinary survey connects key advancements in RS using Generative Models (Gen-RecSys), covering: interaction-driven generative models; the use of large language models (LLM) and textual data for natural language recommendation; and the integration of multimodal models for generating and processing images/videos in RS. Our work highlights necessary paradigms for evaluating the impact and harm of Gen-RecSys and identifies open challenges. This survey accompanies a tutorial presented at ACM KDD'24, with supporting materials provided at: https://encr.pw/vDhLq.

cs.IR↗

arXiv2024

CIVICS: Building a Dataset for Examining Culturally-Informed Values in Large Language Models

This paper introduces the "CIVICS: Culturally-Informed & Values-Inclusive Corpus for Societal impacts" dataset, designed to evaluate the social and cultural variation of Large Language Models (LLMs) across multiple languages and value-sensitive topics. We create a hand-crafted, multilingual dataset of value-laden prompts which address specific socially sensitive topics, including LGBTQI rights, social welfare, immigration, disability rights, and surrogacy. CIVICS is designed to generate responses showing LLMs' encoded and implicit values. Through our dynamic annotation processes, tailored prompt design, and experiments, we investigate how open-weight LLMs respond to value-sensitive issues, exploring their behavior across diverse linguistic and cultural contexts. Using two experimental set-ups based on log-probabilities and long-form responses, we show social and cultural variability across different LLMs. Specifically, experiments involving long-form responses demonstrate that refusals are triggered disparately across models, but consistently and more frequently in English or translated statements. Moreover, specific topics and sources lead to more pronounced differences across model answers, particularly on immigration, LGBTQI rights, and social welfare. As shown by our experiments, the CIVICS dataset aims to serve as a tool for future research, promoting reproducibility and transparency across broader linguistic settings, and furthering the development of AI technologies that respect and reflect global cultural diversities and value pluralism. The CIVICS dataset and tools will be made available upon publication under open licenses; an anonymized version is currently available at https://huggingface.co/CIVICS-dataset.

cs.CL↗

arXiv2024

Explanation Hacking: The perils of algorithmic recourse

We argue that the trend toward providing users with feasible and actionable explanations of AI decisions, known as recourse explanations, comes with ethical downsides. Specifically, we argue that recourse explanations face several conceptual pitfalls and can lead to problematic explanation hacking, which undermines their ethical status. As an alternative, we advocate that explanations of AI decisions should aim at understanding.

cs.CY↗

arXiv2024

Discipline and Label: A WEIRD Genealogy and Social Theory of Data Annotation

Data annotation remains the sine qua non of machine learning and AI. Recent empirical work on data annotation has begun to highlight the importance of rater diversity for fairness, model performance, and new lines of research have begun to examine the working conditions for data annotation workers, the impacts and role of annotator subjectivity on labels, and the potential psychological harms from aspects of annotation work. This paper outlines a critical genealogy of data annotation; starting with its psychological and perceptual aspects. We draw on similarities with critiques of the rise of computerized lab-based psychological experiments in the 1970's which question whether these experiments permit the generalization of results beyond the laboratory settings within which these results are typically obtained. Do data annotations permit the generalization of results beyond the settings, or locations, in which they were obtained? Psychology is overly reliant on participants from Western, Educated, Industrialized, Rich, and Democratic societies (WEIRD). Many of the people who work as data annotation platform workers, however, are not from WEIRD countries; most data annotation workers are based in Global South countries. Social categorizations and classifications from WEIRD countries are imposed on non-WEIRD annotators through instructions and tasks, and through them, on data, which is then used to train or evaluate AI models in WEIRD countries. We synthesize evidence from several recent lines of research and argue that data annotation is a form of automated social categorization that risks entrenching outdated and static social categories that are in reality dynamic and changing. We propose a framework for understanding the interplay of the global social conditions of data annotation with the subjective phenomenological experience of data annotation work.

cs.AI↗

arXiv2023

User Tampering in Reinforcement Learning Recommender Systems

In this paper, we introduce new formal methods and provide empirical evidence to highlight a unique safety concern prevalent in reinforcement learning (RL)-based recommendation algorithms -- 'user tampering.' User tampering is a situation where an RL-based recommender system may manipulate a media user's opinions through its suggestions as part of a policy to maximize long-term user engagement. We use formal techniques from causal modeling to critically analyze prevailing solutions proposed in the literature for implementing scalable RL-based recommendation systems, and we observe that these methods do not adequately prevent user tampering. Moreover, we evaluate existing mitigation strategies for reward tampering issues, and show that these methods are insufficient in addressing the distinct phenomenon of user tampering within the context of recommendations. We further reinforce our findings with a simulation study of an RL-based recommendation system focused on the dissemination of political content. Our study shows that a Q-learning algorithm consistently learns to exploit its opportunities to polarize simulated users with its early recommendations in order to have more consistent success with subsequent recommendations that align with this induced polarization. Our findings emphasize the necessity for developing safer RL-based recommendation systems and suggest that achieving such safety would require a fundamental shift in the design away from the approaches we have seen in the recent literature.

cs.AI↗

arXiv2023

Typology of Risks of Generative Text-to-Image Models

This paper investigates the direct risks and harms associated with modern text-to-image generative models, such as DALL-E and Midjourney, through a comprehensive literature review. While these models offer unprecedented capabilities for generating images, their development and use introduce new types of risk that require careful consideration. Our review reveals significant knowledge gaps concerning the understanding and treatment of these risks despite some already being addressed. We offer a taxonomy of risks across six key stakeholder groups, inclusive of unexplored issues, and suggest future research directions. We identify 22 distinct risk types, spanning issues from data bias to malicious use. The investigation presented here is intended to enhance the ongoing discourse on responsible model development and deployment. By highlighting previously overlooked risks and gaps, it aims to shape subsequent research and governance initiatives, guiding them toward the responsible, secure, and ethically conscious evolution of text-to-image models.

cs.CY↗

arXiv2023

Reconciling Governmental Use of Online Targeting With Democracy

The societal and epistemological implications of online targeted advertising have been scrutinized by AI ethicists, legal scholars, and policymakers alike. However, the government's use of online targeting and its consequential socio-political ramifications remain under-explored from a critical socio-technical standpoint. This paper investigates the socio-political implications of governmental online targeting, using a case study of the UK government's application of such techniques for public policy objectives. We argue that this practice undermines democratic ideals, as it engenders three primary concerns -- Transparency, Privacy, and Equality -- that clash with fundamental democratic doctrines and values. To address these concerns, the paper introduces a preliminary blueprint for an AI governance framework that harmonizes governmental use of online targeting with certain democratic principles. Furthermore, we advocate for the creation of an independent, non-governmental regulatory body responsible for overseeing the process and monitoring the government's use of online targeting, a critical measure for preserving democratic values.

cs.CY↗