AI ethics on Saturday, 4 July 2026

For the past 14 years, Spokane residents learned about upcoming sales and promotions at locally owned Fred's Appliance through the voice and smiling face of Amber George. But in February, George noticed on social media that Fred's, and pro ... (https://incidentdatabase.ai/cite/1565#7484)

Indore: A local resident was defrauded of approximately Rs 1.80 lakh by cybercriminals who used AI-generated voice to impersonate his relative living abroad. "The Indore crime branch has launched an investigation into the case after the vi ... (https://incidentdatabase.ai/cite/1566#7485)

WEST TOWN — In what’s either a mea culpa, a bit of clever marketing or maybe both, a company whose food delivery robot smashed through the glass at a West Town bus shelter last month is now running an apology ad — at the very same bus shelt ... (https://incidentdatabase.ai/cite/1567#7486)

OLD TOWN — For the second time in a week, a self-driving food delivery robot has crashed into a CTA bus shelter, sending shards of glass all over the sidewalk. A Coco robot collided with the glass at a bus shelter about 4 p.m. Tuesday at t ... (https://incidentdatabase.ai/cite/1568#7487)

July 1 (Reuters) - A California man sued OpenAI and its CEO Sam Altman on Wednesday, claiming the company's ChatGPT platform exacerbated his bipolar disorder due to a lack of safeguards for users with mental illness. Michael Lines, 34, sai ... (https://incidentdatabase.ai/cite/1569#7488)

Led by Florida’s James Uthmeier, 27 state attorneys general filed an amicus brief before the Supreme Court supporting a Texas online safety law.

America is a uniquely sick, unhealthy country — just ask Americans. We’re addicted to ultraprocessed food and succumb to deaths of despair. The current US health secretary, who insists we’ve been raising the “sickest generation” ever, has built an entire political movement around the idea that there is something uniquely unwell about America as a […]

If your week flew by — we know ours did — catch up here with what you might have missed. The week at Retraction Watch featured: In case you missed the news, the Hijacked Journal Checker now has more than 450 entries. The Retraction Watch Database has over 65,000 retractions. Our list of COVID-19 retractions … Continue reading Weekend reads: Taylor Swift teaches botany; NEJM retracts key study in Amgen drug; hidden prompts at conference ‘snare AI peer reviews’

Joint statement on AI-generated or altered child sexual exploitation or sexual abuse material by the Cybercrime Convention Committee (T-CY) and the Committee of the Parties to the Convention on the ...

Bangladesh has an estimated 1.17 mental-health professionals per 100,000 population and only six child psychiatrists nationwide. No Bengali-language, culturally adapted tool exists for early screening of abuse-related psychological trauma in children. We present ShishuRaksha AI, a decision-support (not diagnostic) framework that fuses four screening modalities: validated questionnaires (SDQ, CPSS), Bengali narrative text, House-Tree-Person (HTP) drawing features, and facial affect. The fusion is

Low-dimensional projections support interactive visual analysis of high-dimensional data embeddings, but their structure often does not align with analyst-defined semantic relationships. Recent LLM-augmented semantic steering methods address this gap by externalizing analyst intent from user-defined groups of seed examples, but they propagate intent through per-item LLM reasoning, causing LLM calls and cost to grow linearly with collection size. We propose a scalable semantic steering method tha

Sponsored search plays a crucial role as a revenue stream for search engines, wherein advertisers competitively bid on keywords that align with the users' search queries. The task of matching relevant keywords to these queries is complicated by the vast and ever-evolving space of keywords, the ambiguity of user and advertiser intentions, and the wide range of topics and languages involved. Consequently, ensuring that ads are pertinent to user queries presents significant challenges. In the fast-

Sponsored search plays a crucial role in e-commerce revenue generation, where advertisers strategically bid on keywords to capture the attention of users through relevant search queries. However, the process of identifying pertinent keywords for a given query presents significant challenges because of a vast and evolving keyword landscape, ambiguous intentions, and topic diversity. This paper highlights an opportunity for to earn a considerable amount of Ads revenue and user engagement where a s

Sleep physiology arises from the coordinated dynamics of the central nervous system (CNS) and autonomic nervous system (ANS), as reflected by multimodal polysomnography signals including EEG, EOG, EMG, ECG, and respiration. However, existing sleep foundation models often fuse heterogeneous biosignals in a topology-agnostic manner, overlooking their physiological organization. We introduce Omni-Sleep, a sleep foundation model that uses the CNS/ANS partition as a physiological prior for topology-c

Generative models such as diffusion and flow matching have advanced robotic visuomotor policies by modeling multimodal action distributions, but their multi-step sampling or ODE solving introduces inference latency. Existing one-step acceleration methods often compress the whole generation process into a single large update, leading to spatial deviation, frequency distortion, and mode averaging. This paper proposes a high-fidelity one-step generative visuomotor policy framework that addresses th

Despite remarkable progress in machine translation, Vision Language Models (VLMs) struggle on historical manuscripts, a domain that stresses core Natural Language Processing (NLP) capabilities: low-resource transliteration, archaic vocabulary, and noisy input signals. We present a systematic framework for evaluating the full image-to-translation pipeline on medieval Latin manuscripts, a setting in which scribal shorthand, ligatures, and parchment degradation expose failure modes that are invisib

Challenges remain in ego-centric 3D scene generation due to limited view overlap and the dominant influence of individual perspectives on scene interpretation. These factors hinder the creation of viewpoint-consistent and semantically aligned visual content, as well as the construction of accurate geometric structures. In this paper, we propose CGGS, a text-to-3D framework aiming to enhance 3D-content-awareness and address geometric distortions in ego-centric scene generation. Firstly, the Ego-c

Audio foundation models are widely adopted as general-purpose feature extractors, yet the internal structure of their learned representations remains insufficiently understood. In this work, we analyze CLAP audio embeddings through a probing framework, studying the encoding of three fundamental perceptual dimensions: reverberation (RT60), loudness (LUFS), and spectral content, measured via spectral centroid (SC) and relative pitch (RP). Probes of increasing complexity are trained to predict each

Reinforcement Learning (RL) has emerged as a powerful paradigm for adaptive traffic signal control. However, in safety-critical infrastructure like traffic control, the opaque, black-box nature of deep RL models poses challenges for transportation agency acceptance, regulatory compliance, operational trust, troubleshooting, and fine-tuning. To bridge this gap between high-performance optimization and human-comprehensible interpretability, this effort introduces a novel, explainable entity centri

Relational databases (RDBs) are the primary data infrastructure in many enterprises, yet recent deep learning methods designed for RDBs have been evaluated under inconsistent experimental protocols, making fair comparison difficult. We present one of the first systematic benchmarking studies of recently released deep learning methods for RDBs, evaluating them across five relational databases, with one classification and one regression task for each. We refactor all deep RDB models to allow the f

Plasma diagnostic models for tokamak fusion devices are almost universally evaluated on clean, complete sensor data. In practice, fusion diagnostics fail regularly: acquisition systems start late, individual sensors die, and signal dropouts cluster precisely when a plasma disruption is approaching. We present the first systematic robustness benchmark for plasma diagnostic ML using the TokaMark dataset of 11,573 MAST shots, evaluating XGBoost, LSTM, Transformer, and the TokaMark CNN baseline acro

Social Media + Society, Volume 12, Issue 3, July-September 2026. This study examines how conspiracy communities on Reddit perceive and critique algorithms, emphasizing the need to integrate individual and socially centered approaches to understand algorithm-driven social automation more broadly. As algorithms ...

Visual object tracking is a fundamental problem in computer vision. Traditional tracking methods, which primarily rely on RGB imagery, often face difficulties in complex scenarios such as low resolution and background clutter. Hyperspectral imaging, which captures both spatial and spectral information across multiple narrow spectral bands, has emerged as a promising solution. However, hyperspectral tracking suffers from challenges including the complexity of spatial-spectral-temporal modeling, t

Large language models are increasingly deployed as IDE-integrated coding agents that decompose tasks, generate and edit files, run code, and refine outputs over many turns. Yet their safety is still often evaluated as if they were chatbots: one harmful prompt, one response, judged in isolation. We introduce workflow-level jailbreak construction, a failure mode in which a harmful objective is assembled across ordinary stages of a software-development workflow rather than generated through a singl

Delivering a sustainable future – for our environment, energy supply, businesses, health, and social institutions – is a global challenge. To meet this challenge, LSE and Tsinghua University have ...

In chest X-ray (CXR) classification, acceptable ranking performance can still leave rare-positive patients below threshold, especially within subgroups. We study this pre-deployment fairness problem as an audit question: after a long-tailed multi-label CXR model is converted from scores into decisions, who is missed? Across VinDr-CXR and MIMIC-CXR/CXR-LT, we use a diagnostic ladder to separate class-level long-tail losses, subgroup-aware weighting, group robustness, and threshold selection. On V