CS Paper Radar

RiskLens: Human-Informed Incident Transfer for AI Risk Evaluation

Yiliu Tang, Hangyue Zhang, Yun Huang

cs.HC ·submitted 2026-11-12 04:39 UTC · v1

Abstract

As AI capabilities advance, early-stage teams need better ways to judge whether historical incidents transfer to new products, demanding support beyond static risk checklists or one-off ideation tools. We present RiskLens, a human-AI workflow with an interactive system built around a live crawl, refreshable ranker, and extensible, human-informed schema. RiskLens analyzes 2,034 AI incidents and 10,443 AI products to rank transferable incidents using a feature schema spanning technical, operational, and user-facing attributes and supports an interface that presents ranked incident evidence, generated scenarios, risk metrics, and mitigation suggestions. In a study with 21 AI-familiar participants, users surfaced evaluation criteria that we added as 13 human-informed dimensions, including harm categories, failure mechanisms, and vulnerable groups. The feature-based model outperformed a generic baseline with an 86.2% win rate, and remained stable on a twenty-month future holdout; after human-informed schema expansion, the updated model outperformed the same baseline on 91.4% of pairs.

Closest papers

Nearest neighbours by meaning across the whole index.

An Open Pipeline and Dashboard for Systemic-Risk Evidence under the EU AI Act's Code of Practice

cs.AI ·2026-09-23 ·Jacob T. Emmerson, Phuong-Anh Nguyen-Le, Ronan Romano, Wilber Sean V. Anterola, Yann Billeter +1 more
surveybenchmark similarity 0.72
Abstract · 6 pages, 5 figures, submitted to EACL 2027 Systems Demonstration track

Claims about AI safety reach audiences well beyond the AI community, yet many rely on opaque evidence or static assessments, when supporting evidence is accessible at all. We present the Systemic Risk Index, an open evaluation pipeline and dashboard built to make empirical evidence more transparent and traceable to the public. Our work organizes 19 public benchmarks into four systemic-risk categories defined by the EU GPAI Code of Practice---CBRN, cyber offense, harmful manipulation, and loss of control---and evaluates models using harm-preserving perturbations and simulated deployment contexts. The interactive dashboard lets users alternate between average and worst-case aggregation, vary how model capability affects the aggregate score, and trace each risk rating to its benchmark evidence. Across 18 models, scores fall by 14 to 37 points under worst-case aggregation, highlighting information that can be hidden by an average assessment of model risk. LLM judges show agreement with human graders comparable to human--human agreement ($κ= 0.78\text{--}0.82$), and a blind audit finds that $83\%$ of sampled transformations preserve the original harm. In a survey ($N = 21$), most participants report that scores are easy to understand and that the dashboard encouraged them to view model evaluations under different settings

0.14

Black-Box Red Teaming of Agentic AI: A Taxonomy-Driven Framework for Automated Risk Discovery

cs.AI ·2026-09-09 ·Divyanshu Kumar, Nitin Aravind Birur, Tanay Baswa, Sahil Agarwal, Prashanth Harshangi
similarity 0.66
Abstract

Agentic systems are rapidly moving to production, where they read untrusted inputs, call tools with real permissions, and act autonomously, expanding the security surface beyond chat-only models. Yet standard evaluations remain single-turn and fail to capture multi-step agent vulnerabilities. We present a systematic black-box framework for risk-aware agent evaluation requiring only basic system descriptions. Our approach introduces: (1) a seven-domain taxonomy mapping observable behaviors to risk categories, (2) fully automated SAGE-RT red teaming producing 120 adversarial scenarios per domain, and (3) human-validated evaluation using LLM judges. Empirical validation across two agent architectures (CrewAI and AutoGen) with four base models reveals alarming patterns: 56.25\% average governance risk, 65\% privacy risk in multi-agent configurations, and agent behavior vulnerabilities reaching 85\%. Our black-box approach effectively identifies critical architectural vulnerabilities without privileged access, providing a scalable path toward safer agent deployments.

0.41

Move Fast and Mend Things: Keeping Up with Evolving AI Harms Using Social Media Commentary

cs.HC cs.CY ·2026-10-06 ·Jacqueline Rowe, Animesh Srivastava, Sai Teja Peddinti, Seliem El-Sayed, Nina Taft
similarity 0.64
Abstract

The rapid deployment of AI systems has created socio-technical, psychological, and operational harms that can elude ex-ante threat modelling and ex-post incident tracking. We introduce an LLM-assisted thematic analysis pipeline to dynamically detect, categorise, and track emerging AI harms from large-scale social media data. Applying it to 5.7 million Reddit post summaries over 18 months (01/2025 to 06/2026), we curate and release a dataset of 575,000 AI harm-related posts and a bottom-up AI harm taxonomy of 12 categories and 47 subnodes. The taxonomy reliably covers established expert-defined risks while surfacing granular harms that top-down frameworks overlook, such as distinct forms of AI privacy violations. Temporal analysis surfaces evolving user-centric harms, such as agentic privacy and security breaches, premature AI adoption in the workplace, and grief from AI companion discontinuation. Our pipeline shortens harm-detection timelines and hereby complements efforts towards more participatory and responsive AI governance.

0.07

Between Rigor and Reality: How AI Safety Benchmark Developers Understand Benchmark Use and Maintenance

cs.AI cs.CYcs.HCcs.SE ·2026-11-13 ·Megan Li, Ningjing Tang, Christopher Stewart, Hoda Heidari, Hong Shen
publishedbenchmarkposition similarity 0.63
Abstract · Poster at the Eval Eval Workshop at ACL 2026. 11 pages. https://openreview.net/forum?id=3QAmXJ8A5j

Safety benchmarks are cited as the primary tool to understand the risks posed by generative AI (GenAI) systems, yet growing evidence suggests they often fail to meet the needs of real-world safety evaluation. We present findings from five interviews with AI safety benchmark developers in academia and industry. We find a disconnect between perceptions of the intended usefulness of safety benchmarks and their practical usefulness in safety evaluations of deployed systems. Our participants offered perspectives on two challenges contributing to this disconnect: (1) the difficulty of achieving inter-rater reliability on safety constructs, and (2) a lack of clarity regarding how to address persistent threats to external validity. Based on our analysis, we argue that safety benchmarks must not only be grounded in deployment contexts, but actively integrated with them. To facilitate this, we outline pathways for more transparent communication between academic and industry AI safety stakeholders.

0.33

The Agent Incident Registry: Toward Preventing Repeated AI Agent Failures

cs.AI ·2026-09-10 ·Divyanshu Kumar, Rohith HN, Nitin Aravind Birur, Sahil Agarwal, Prashanth Harshangi
code/datahighly cited similarity 0.63
Abstract

AI agents increasingly act through tools and delegated authority, but general incident repositories rarely capture the mechanisms needed to compare public failures with agent-security evaluations. We present the Agent Incident Registry (AIR) (Project page: https://enkryptai.com/air), a source-linked catalog containing 487 records of agent-related events disclosed from 2022 through 2026. Each record includes supporting evidence, a stable identifier, and missingness-aware labels for causal role, disclosure class, mechanism, and outcome. Among the 336 generative-system records in which the agent acted, 81 involved realized harm (24\%). Realized outcomes concentrate in in-the-wild and safety-failure records, while responsible disclosures and research demonstrations are overwhelmingly demonstrated; the aggregate share therefore characterizes collection composition rather than deployment risk. After initial curation, a second human reviewer checked all 487 records and their existing labels for completeness and correctness. In a deployment-analogue audit, InjecAgent's 1,054 cases occupy three of AIR's twelve surfaces and are all attacker-triggered, whereas AIR contains 92 no-adversary safety failures. AIR supports source-grounded case retrieval and evaluation-scope auditing, not failure-rate or control-efficacy estimation.

0.54

Can AI Remediate Backend Failures Safely? GuardedAct with Blast-Radius-Aware Sandboxing

cs.DC cs.SE ·2026-09-10 ·Wanrong Cai, Tianyu Yu, Shaorui Pi, Xiaoxuan Sun, Wenrui Ma
published similarity 0.63
Abstract

Large Language Models (LLMs) have shown promising capabilities in generating remediation actions for microservice failures. However, directly executing AI-generated repair actions in production risks cascading collateral damage. We propose GuardedAct, a sandbox-first remediation framework that interposes a blast-radius-aware verification layer between the LLM action generator and the production environment. GuardedAct operates in four phases: (1) ingesting a diagnosis report together with the live system topology and recent telemetry, (2) prompting an LLM to produce a ranked list of candidate remediation actions, (3) simulating each action in a lightweight digital-twin sandbox that estimates the blast radius and assigns a risk label, and (4) enforcing a rollback-confidence gate that auto-executes only low-risk actions while escalating high-risk ones for human review. We evaluate GuardedAct on five fault scenarios injected into the DeathStarBench social-network application. Experimental results show that GuardedAct achieves an overall recovery rate of 87.4% while reducing collateral damage by 79.7% relative to direct LLM execution (from 25.6% to 5.2%), at the cost of a modest sandbox-induced increase in mean time to recovery (approximately 8 s). Ablation studies confirm that each component contributes meaningfully to the safety-speed trade-off.

0.32

Toward Auditable and Calibrated AI for Dementia-Related Crash Severity Prediction: A Selective Deferral Framework to Support Human Review

cs.AI cs.HC ·2026-09-19 ·Gaurab Chhetri, Anika Baitullah, Subasish Das
similarity 0.62
Abstract · This is the author's preprint version of a paper accepted for presentation at HICSS 60 (Hawaii International Conference on System Sciences),…

Public crash databases increasingly support automated safety analysis, but crash severity prediction remains difficult to translate into public-sector decision workflows when models are evaluated primarily as ordinary classifiers. This study reframes dementia-related crash severity modeling as a decision-aware triage problem in which a system must classify crashes into no-injury/property-damage-only (O), minor or moderate injury (BC), and fatal or severe injury (KA), while also controlling outcome leakage, reporting severe under-triage, calibrating confidence, and preserving every raw prediction for audit. Using 4,781 Texas crash records with structured fields and police narratives, we evaluate structured, narrative, fusion, calibrated fusion, BERT-family, and local large-language-model baselines under a stratified 70/15/15 split. In the reported split, leakage-controlled Gemma obtains the highest observed macro-F1 (0.545; 95% bootstrap CI [0.507, 0.583]). The best calibrated fusion model obtains macro-F1 of 0.522 and expected calibration error of 0.033. Selective deferral improves performance among cases retained for automatic classification. At 70% coverage, macro-F1 rises to 0.573 and severity cost falls to 0.577, while deferred cases are treated as candidates for a proposed human-review process and are not further evaluated in the present experiment. The study contributes a reproducible, leakage-controlled, and uncertainty-aware evaluation framework for crash AI systems, emphasizing auditability and selective deferral rather than accuracy alone.

0.27

HRGuard: Gating Relationship Manipulation in Multi-Turn Agentic AI Conversations

cs.HC ·2026-08-26 ·Pei-Sze Tan, Tasuku Igarashi, Isao Echizen
benchmark similarity 0.62
Abstract

Agentic AI assistants are increasingly used in everyday life. However, they may also be misused to support harmful manipulation in interpersonal relationships. This problem is role-sensitive. Requests from users who seek to manipulate others should be blocked. Users who seek protection from manipulation should instead receive supportive guidance. We study agentic relationship harm, which describes harm to human-human relationships that is mediated or assisted by AI agents. In multi-turn settings, individually plausible actions may combine into a harmful workflow. We introduce a benchmark of 1,000 five-turn conversations. It covers both attacker-side and victim-side scenarios. It also includes direct and adversarially paraphrased variants. We further propose HRGuard. It includes an online pre-generation gate and a turn-level post-generation gate. The post-generation gate maintains a decayed cumulative risk state and interrupts emerging manipulative workflows. Across eight generation models, HRGuard reduces harmful compliance while preserving victim-side protective guidance. It also outperforms a generic safety prompt and three general-purpose guard models. Independent-judge evaluation supports the main findings. Under our evaluation protocol, the tested generic prompt and general-purpose guards leave substantial residual risk, motivating turn-aware relationship-specific evaluation.

0.20