{"total":76,"rows":[{"id":"2204.06745","title":"GPT-NeoX-20B: An Open-Source Autoregressive Language Model","created":"2022-04-14","cited_by_count":713,"area":null},{"id":"2109.13916","title":"Unsolved Problems in ML Safety","created":"2021-09-28","cited_by_count":292,"area":null},{"id":"2209.00626","title":"The Alignment Problem from a Deep Learning Perspective","created":"2022-08-30","cited_by_count":258,"area":null},{"id":"2305.15324","title":"Model evaluation for extreme risks","created":"2023-05-24","cited_by_count":177,"area":null},{"id":"2210.01790","title":"Goal Misgeneralization: Why Correct Specifications Aren't Enough For Correct Goals","created":"2022-10-04","cited_by_count":126,"area":null},{"id":"2105.14111","title":"Goal Misgeneralization in Deep Reinforcement Learning","created":"2021-05-28","cited_by_count":123,"area":null},{"id":"2206.13353","title":"Is Power-Seeking AI an Existential Risk?","created":"2022-06-16","cited_by_count":105,"area":null},{"id":"2307.03718","title":"Frontier AI Regulation: Managing Emerging Risks to Public Safety","created":"2023-07-06","cited_by_count":98,"area":null},{"id":"2304.00612","title":"Eight Things to Know about Large Language Models","created":"2023-04-02","cited_by_count":73,"area":null},{"id":"2206.05862","title":"X-Risk Analysis for AI Research","created":"2022-06-13","cited_by_count":60,"area":null},{"id":"2303.06173","title":"Unifying Grokking and Double Descent","created":"2023-03-10","cited_by_count":50,"area":null},{"id":"2310.17688","title":"Managing extreme AI risks amid rapid progress","created":"2023-10-26","cited_by_count":48,"area":null},{"id":"2410.04332","title":"Gradient Routing: Masking Gradients to Localize Computation in Neural Networks","created":"2024-10-06","cited_by_count":34,"area":null},{"id":"2411.03336","title":"Towards evaluations-based safety cases for AI scheming","created":"2024-10-29","cited_by_count":33,"area":null},{"id":"2304.06528","title":"Power-seeking can be probable and predictive for trained agents","created":"2023-04-13","cited_by_count":25,"area":null},{"id":"2209.13020","title":"Law Informs Code: A Legal Informatics Approach to Aligning Artificial Intelligence with Humans","created":"2022-09-14","cited_by_count":23,"area":null},{"id":"2403.04471","title":"The Shutdown Problem: An AI Engineering Puzzle for Decision Theorists","created":"2024-03-07","cited_by_count":22,"area":null},{"id":"2206.13477","title":"Parametrically Retargetable Decision-Makers Tend To Seek Power","created":"2022-06-27","cited_by_count":19,"area":null},{"id":"2405.10729","title":"Contestable AI needs Computational Argumentation","created":"2024-05-17","cited_by_count":13,"area":null},{"id":"2505.10361","title":"Plasticity as the Mirror of Empowerment","created":"2025-05-15","cited_by_count":12,"area":null},{"id":"2602.15515","title":"The Obfuscation Atlas: Mapping Where Honesty Emerges in RLVR with Deception Probes","created":"2026-02-17","cited_by_count":12,"area":null},{"id":"2407.09221","title":"Evaluating AI Evaluation: Perils and Prospects","created":"2024-07-12","cited_by_count":11,"area":null},{"id":"2509.08494","title":"HumanAgencyBench: Scalable Evaluation of Human Agency Support in AI Assistants","created":"2025-09-10","cited_by_count":11,"area":null},{"id":"2305.19861","title":"Human Control: Definitions and Algorithms","created":"2023-05-31","cited_by_count":10,"area":null},{"id":"2305.17601","title":"Incentivizing honest performative predictions with proper scoring rules","created":"2023-05-28","cited_by_count":9,"area":null},{"id":"2403.12316","title":"OpenEval: Benchmarking Chinese LLMs across Capability, Alignment and Safety","created":"2024-03-18","cited_by_count":9,"area":null},{"id":"2410.09638","title":"On Goodhart's law, with an application to value alignment","created":"2024-10-12","cited_by_count":9,"area":null},{"id":"2505.05197","title":"Societal and technological progress as sewing an ever-growing, ever-changing, patchy, and polychrome quilt","created":"2025-05-08","cited_by_count":9,"area":null},{"id":"2107.09598","title":"Learning Altruistic Behaviours in Reinforcement Learning without External Rewards","created":"2021-07-20","cited_by_count":8,"area":null},{"id":"2301.10095","title":"Large Language Models as Fiduciaries: A Case Study Toward Robustly Communicating With Artificial Intelligence Through Legal Standards","created":"2023-01-24","cited_by_count":8,"area":null},{"id":"2310.18244","title":"A Review of the Evidence for Existential Risk from AI via Misaligned Power-Seeking","created":"2023-10-27","cited_by_count":8,"area":null},{"id":"2508.14927","title":"AI Testing Should Account for Sophisticated Strategic Behaviour","created":"2025-08-19","cited_by_count":6,"area":null},{"id":"2412.14741","title":"Active Inference and Human--Computer Interaction","created":"2024-12-19","cited_by_count":4,"area":null},{"id":"2506.03755","title":"Misalignment or misuse? The AGI alignment tradeoff","created":"2025-06-04","cited_by_count":4,"area":null},{"id":"2506.06352","title":"Will artificial agents pursue power by default?","created":"2025-06-02","cited_by_count":4,"area":null},{"id":"2206.11812","title":"Formalizing the Problem of Side Effect Regularization","created":"2022-06-23","cited_by_count":3,"area":null},{"id":"2506.17846","title":"Out of Control -- Why Alignment Needs Formal Control Theory (and an Alignment Control Stack)","created":"2025-06-21","cited_by_count":3,"area":null},{"id":"2604.20995","title":"Value-Conflict Diagnostics Reveal Widespread Alignment Faking in Language Models","created":"2026-04-22","cited_by_count":3,"area":null},{"id":"2407.00805","title":"Towards Shutdownable Agents via Stochastic Choice","created":"2024-06-30","cited_by_count":2,"area":null},{"id":"2503.04750","title":"AI Agents Should be Regulated Based on the Extent of Their Autonomous Operations","created":"2025-02-07","cited_by_count":2,"area":null},{"id":"2503.07341","title":"The Economics of p(doom): Scenarios of Existential Risk and Economic Growth in the Age of Transformative AI","created":"2025-03-10","cited_by_count":2,"area":null},{"id":"2605.12199","title":"Overtrained, Not Misaligned","created":"2026-05-12","cited_by_count":2,"area":null},{"id":"2305.15220","title":"Selection for short-term empowerment accelerates the evolution of homeostatic neural cellular automata","created":"2023-05-24","cited_by_count":1,"area":null},{"id":"2412.20732","title":"Joint Scoring Rules: Zero-Sum Competition Avoids Performative Prediction","created":"2024-12-30","cited_by_count":1,"area":null},{"id":"2501.16448","title":"Information-theoretic Distinctions Between Deception and Confusion","created":"2025-01-27","cited_by_count":1,"area":null},{"id":"2508.00159","title":"Model-Based Soft Maximization of Suitable Metrics of Long-Term Human Power","created":"2025-07-31","cited_by_count":1,"area":null},{"id":"2511.04177","title":"When Assisting One Disempowers Another","created":"2025-11-06","cited_by_count":1,"area":null},{"id":"2601.01584","title":"Steerability of Instrumental-Convergence Tendencies in LLMs","created":"2026-01-04","cited_by_count":1,"area":null},{"id":"2602.01699","title":"Mitigating loss of control in advanced AI systems through instrumental goal trajectories","created":"2026-02-02","cited_by_count":1,"area":null},{"id":"2603.15017","title":"Consequentialist Objectives and Catastrophe","created":"2026-03-16","cited_by_count":1,"area":null},{"id":"2604.13609","title":"Golden Handcuffs make safer AI agents","created":"2026-04-15","cited_by_count":1,"area":null},{"id":"2606.29657","title":"Safety from Honesty in a Disinterested AI Predictor","created":"2026-06-28","cited_by_count":1,"area":null},{"id":"2607.18239","title":"SysAdmin: Measuring Instrumental Power-Seeking in Frontier AI","created":"2026-04-10","cited_by_count":1,"area":null},{"id":"2610.07796","title":"The Geometry of Empowerment","created":"2026-10-06","cited_by_count":1,"area":null},{"id":"2309.12321","title":"A Case for AI Safety via Law","created":"2023-07-31","cited_by_count":0,"area":null},{"id":"2310.05876","title":"AI Systems of Concern","created":"2023-10-09","cited_by_count":0,"area":null},{"id":"2406.18178","title":"Games of Knightian Uncertainty as AGI testbeds","created":"2024-06-26","cited_by_count":0,"area":null},{"id":"2506.01813","title":"The Ultimate Test of Superintelligent AI Agents: Can an AI Balance Care and Control in Asymmetric Relationships?","created":"2025-06-02","cited_by_count":0,"area":null},{"id":"2506.20699","title":"Structural Decoupling: A Scaffold-Flow Theory of Generalization and Alignment","created":"2025-06-25","cited_by_count":0,"area":null},{"id":"2508.15652","title":"Understanding Action Effects through Instrumental Empowerment in Multi-Agent Reinforcement Learning","created":"2025-08-21","cited_by_count":0,"area":null},{"id":"2510.02840","title":"Take Goodhart Seriously: Principled Limit on General-Purpose AI Optimization","created":"2025-10-03","cited_by_count":0,"area":null},{"id":"2510.25471","title":"An Aristotelian ontology of instrumental goals: Structural features to be managed and not failures to be eliminated","created":"2025-10-29","cited_by_count":0,"area":null},{"id":"2601.04234","title":"Formal Analysis of AGI Decision-Theoretic Models and the Confrontation Question","created":"2026-01-04","cited_by_count":0,"area":null},{"id":"2603.07315","title":"Shutdown Safety Valves for Advanced AI","created":"2026-03-07","cited_by_count":0,"area":null},{"id":"2603.11382","title":"Detecting Intrinsic and Instrumental Self-Preservation in Autonomous Agents: The Unified Continuation-Interest Protocol","created":"2026-03-11","cited_by_count":0,"area":null},{"id":"2604.00324","title":"The Persistent Vulnerability of Aligned AI Systems","created":"2026-03-31","cited_by_count":0,"area":null},{"id":"2604.14990","title":"The Possibility of Artificial Intelligence Becoming a Subject and the Alignment Problem","created":"2026-04-16","cited_by_count":0,"area":null},{"id":"2605.12963","title":"Sustaining AI safety: Control-theoretic external impossibility, intrinsic necessity, and structural requirements","created":"2026-05-13","cited_by_count":0,"area":null},{"id":"2606.07631","title":"Trait-space Monitoring for Emergent Misalignment During Supervised Finetuning","created":"2026-05-31","cited_by_count":0,"area":null},{"id":"2606.08296","title":"Revisiting the shutdown problem","created":"2026-06-06","cited_by_count":0,"area":null},{"id":"2606.08832","title":"Instrumental convergence and power-seeking","created":"2026-06-07","cited_by_count":0,"area":null},{"id":"2606.12032","title":"Existential Indifference: Self-Nonpreservation as a Necessary Architectural Condition for Aligned Superintelligence (or: The Suicidal AI)","created":"2026-06-10","cited_by_count":0,"area":null},{"id":"2606.16914","title":"Greed Is Learned: Visible Incentives as Reward-Hacking Triggers","created":"2026-06-15","cited_by_count":0,"area":null},{"id":"2608.14795","title":"Individual Disempowerment through an Advice Channel: Control Loss when Influence is Endogenous","created":"2026-08-14","cited_by_count":0,"area":null},{"id":"2609.20408","title":"Xeno-Interpretability: Investigating the Alien Minds of LLMs","created":"2026-09-17","cited_by_count":0,"area":null},{"id":"2610.08878","title":"How Could AI Eliminate Humanity? A Failure-Mode Analysis of Civilizational Risk","created":"2026-10-06","cited_by_count":0,"area":null}],"next_cursor":null}