AI Accountability

Follow this topic Ask about this topic

AI accountability studies how people and institutions can answer for AI systems’ decisions, actions, and effects. Approaches include assigning legal and organizational responsibility, documenting data and decisions, auditing models and deployments, and using provenance records, monitoring, and human review to reconstruct what happened. Work increasingly addresses agents and other systems that act across tools and organizations, as well as established settings such as healthcare, government, and research. A central tension is that technical evidence and transparency can make actions easier to trace without ensuring anyone has the authority, responsibility, or means to contest decisions and provide remedies.

Latest papers

157 papers10 papers in the last four weeks, down 50% on the four weeks before. 0.1% of all new papers.

Safe by Benchmark, Unsafe in Practice: Lessons from 1331 Child-Related AI Incidents
Oct 26, 2026Junfeng Jiao, Kevin Chen, Shrestha Mishra +4AI Safety AI Accountability
The University of Texas at Austin, USA · Allen Institute for AI, USA · Stanford University, USA
A Three-Pillar Governance Framework for Safe and Trustworthy AI Agents
Oct 21, 2026Edward C. Cheng, Jeshua Cheng, Alice SiuAI Agent Governance AI Agent Safety
Management Science & Engineering, Stanford University, Stanford, California, United States · AI Laboratory, Inquiryon, Inc., South San Francisco, California, United States · Deliberative Democracy Lab, Stanford University, Stanford, California, United States
Holding Agents Accountable with Verifiable Agent Insurance
Oct 13, 2026Cheng Hua, Taro Tsuchiya, Nicolas Christin +1AI Accountability AI Agent Governance
Carnegie Mellon University
Who is the 'I' in AI? Establishing Decentralized Identity and Accountability for Autonomous Agents
Oct 12, 2026Ahammod Munim Ramil, Khalid Al Kafi, M. A. Abdullahis Sany +4AI Accountability AI Agent Governance
Department of Computer Science and Engineering, BRAC University, Dhaka, Bangladesh · School of Computing, Mathematics and Engineering, Charles Sturt University, Bathurst, New South Wales, Australia
Transparency as Answerability: Role-Dependent Information Flow Modeling
Oct 12, 2026Mattias Brännström, Themis Dimitra Xanthopoulou, Lili JiangAI-Assisted Decision Making AI Accountability
Umeå University, Sweden
TRACE: A Governance Framework for Measuring Explainability Debt in Production AI Systems
Oct 7, 2026Harish Kant PathakAI Governance AI Accountability
Independent Researcher
Auditable Claims about AI Agents
Oct 5, 2026Yue Zhao, Jiate Li, Li Li +4AI Assurance AI Accountability
University of Southern California, Los Angeles, CA, USA · Arizona State University, Tempe, AZ, USA
The Reported Engagement with AI Level (REAL) Rating: A Framework for Disclosing Human-AI Collaboration
Oct 2, 2026Imène Goumiri, Mayleen Cortez-Rodriguez, Eric Bell +1AI Use Disclosure Human-AI Collaboration
Real Good AI
Cybernetic and Epistemic: A Missing Vocabulary for Trustworthy Agentic Delegation
Oct 1, 2026Jérémie LumbrosoAI Accountability Agentic Code Generation
University of Pennsylvania
A Reusable Semantic Web Framework for Evidence-Grounded Fundamental Rights Impact Assessments under the EU AI Act
Sep 30, 2026Faith Olopade, Delaram Golpayegani, David LewisAI Accountability AI Governance
Trinity College Dublin, Ireland · ADAPT Centre, Trinity College Dublin, Ireland
Policy-Conditioned AI-Use Detection: An Evidentiary Framework for Academic Publishing
Sep 29, 2026Jairo Diaz-Rodriguez, Mumin JiaAI Accountability Scholarly Peer Review
Department of Mathematics and Statistics, York University, Toronto, Ontario M3J 1P3
Beyond the Last Truffula Tree: SustainAI - A Water-Aware, Closed-Loop Framework for Environmentally Accountable AI
Sep 25, 2026Farnaz Farid, Tashfia Towkee, Sania Nasreen +1Responsible AI AI Accountability
Western Sydney University · Southwest Jiaotong University
When No One Owns the Judgment: Accountability Under Contribution Dissolution in Human-AI Collaboration
Sep 24, 2026Hengzhi YeAI Accountability Human-AI Collaboration
School of Computer Science Peking university Beijing, China
"We'll Fix It Later": Education, AI, and the Deferral of Privacy in EdTech
Sep 23, 2026Meghna Manoj Nair, Rachel GreenstadtEducational Technology AI Governance
New York University (United States)
The Disciplinary Language Transfer Problem: How Psychological Vocabulary Produces Governance Failures in AI Agent Deployment
Sep 22, 2026Kymberly Lasser-Chere, Tyler Akidau, Marc MillstoneAI Accountability AI Agent Auditing
Coastal Waters Healing Center, Surf City, NC, USA · Redpanda, Seattle, WA, USA
The Ethics of Artificial Intelligence in Military Operations
Sep 22, 2026Nicolas Drapier, Florian Mauberger, Aladine Chetouani +1AI Accountability AI Governance
L2TI Laboratory, Universit´e Sorbonne Paris Nord, 99 Avenue Jean-Baptiste Cl´ement, Villetaneuse, 93430, France. · SAS Impact, 1 Rue Sainte-Anne, Orl´eans, 45000, France.
Making AI-Assisted Claims Independently Challengeable: Publication Authority and a Protocol for Falsifiable Publication Records
Sep 15, 2026Torsten Olivi Tiltack, Yifei Dong, Kun Yu +5AI Assurance AI Accountability
Data Science Institute, University of Technology Sydney, Ultimo, NSW, Australia · Global Big Data Technologies Centre, University of Technology Sydney, Ultimo, NSW, Australia · School of Computer Science, University of Technology Sydney, Ultimo, NSW, Australia
AI Deployment Accountability Engineering: A Vision for Accountable AI in Safety-Critical Socio-Technical Systems
Sep 13, 2026Murat KantarciogluAI Accountability AI Risk Management
Virginia Tech Blacksburg, VA, USA
From Protocols to Evidence: Bounded Claims for AI in Service of the Common Good
Sep 10, 2026Nitesh V. Chawla, Paulo BenantiAI Accountability AI Safety Evaluation
University of Notre Dame Notre Dame, USA · LUISS Guido Carli University Rome, Italy
SoulAuth: An Actor-native Identity Architecture and Rust Reference Implementation for Humans and Long-lived AI Actors
Sep 10, 2026Kun Yuan, Harold Wang, Echo Li +4AI Accountability Cybersecurity
TRANTOR LABS, Singapore
AI Safety: Not Optional, Not Later
Sep 9, 2026Qinghua Lu, Yoshua BengioAI Assurance AI Accountability
CSIRO, Australia · LawZero, Canada · Université de Montréal, Canada +1
Watermarks Without Verification: AI Text Watermarking After the EU AI Act
Sep 9, 2026Alexander Nemecek, Vipin Chaudhary, Erman AydayText Watermarking AI Accountability
Case Western Reserve University
Accountable and uncertainty-aware evaluation of sensor-based AI under distribution shift: devices, subjects, and nearly three years underground
Sep 8, 2026Benny Platte, Rico Thomanek, Christian Roschke +1AI Accountability Uncertainty Quantification
Mittweida University of Applied Sciences
Silent Revision: Measuring Undisclosed Change in the Safety Frameworks of Frontier AI Developers
Sep 8, 2026Louis Yiven ZhuAI Accountability AI Safety
University of Oxford
Audit Without Verification: When LLM Accountability Layers Relay Rather Than Check
Sep 7, 2026Paul-Peter ArslanAgent Failure Analysis AI Accountability
Institute For Future Technologies
A Unified Policy Architecture (UPA): The Governance Kernel for Enterprise AI Operating Systems
Sep 6, 2026Prabhu Raghav, Balamurugan Pandi, Arul Vivek +2AI Agent Governance AI Governance
SuperAgentX AI
Measuring AI Accountability Through Argumentation Analysis: Can Model Reasoning Withstand Scrutiny?
Sep 4, 2026Daan R. Henselmans, Derck W. E. Prinzhorn, Arno LibertAI Accountability Language Model Auditing
Aithos Research Foundation, Keizersgracht 62, Amsterdam, 1015 CS, the Netherlands.
When Does an Interpretation Count as Established? The Formation, Evaluation, and Responsibility of Interpretation in Generative AI
Sep 4, 2026Deyu JingAI Accountability Generative AI Evaluation
Fudan University, Shanghai, China
Accountable AI with Grounded, Faithful, Consistent, Actionable Rationales: A Case Study in Clinical Trial Matching with VERDICT
Sep 3, 2026Zikai Zhou, Yufei Jin, Yilin Xu +3AI Accountability LLM Decision-Making
Department of Computer Science, Stanford University, Stanford, CA, USA · Samueli Electrical and Computer Engineering, UCLA, Los Angeles, CA, USA · Department of Computer Science, Emory University, Atlanta, GA, USA +1
A Mathematical Framework for Legacy, Governance, and Decision Integrity in Enterprise AI
Sep 1, 2026Shorab SarkerAI-Assisted Decision Making AI Accountability
Independent Researcher · The Math Behind Innovation

Oct 26, 2026J/K move · Enter open · S save

Safe by Benchmark, Unsafe in Practice: Lessons from 1331 Child-Related AI Incidents

Junfeng Jiao, Kevin Chen, Shrestha Mishra, Abhejay Murali+3

The University of Texas at Austin, USA · Allen Institute for AI, USA · Stanford University, USA

AI Safety AI Accountability

Open paper AskPDF arXiv

AI systems are increasingly present in children's everyday environments: education, entertainment, social networks, and toys. Their safety is typically assessed at the model level, through benchmarks, red-teaming, and content filtering. However, most AI harm taxonomies lack child-specific evidence and under-specify developmental harm and the stakeholder asymmetries unique to minors. Building on prior taxonomies, we construct a child-centered harm taxonomy mapped to sociotechnical failure modes. We curate and code 1331 publicly reported child-related AI incidents, annotating internal and external harm types and the failure mode implicated in each incident. Our analysis reveals which harms are most prevalent in public reporting, how individual-level and population-level harms relate, and which failures are most often coded alongside harm. We find that harm concentrates in representational, dignity-related, and psychological categories, and co-occurs most often with deployment-context, governance, and misuse failures rather than model defects. Harm to an individual child is also frequently linked to population-level harm, reaching children who never interacted with the system. These findings suggest that model-level evaluations, while necessary, cannot by themselves anticipate the harm pathways most often documented for children. We conclude with actionable implications emphasizing mitigations beyond model-level safeguards such as content filtering: age assurance, consent verification for depiction, safety-by-design UX constraints, incident disclosure norms, and auditing regimes targeting developmental risk. Privacy-minimized annotations, a codebook, and documentation are available at https://github.com/The-Responsible-AI-Initiative/KidAID.