AI ethics on Monday, 6 July 2026

WASHINGTON - Maria believed she was romancing a prince from Dubai, captivated by his flirtatious smile and declarations of affection he showered on her during live video calls. But the suitor was an AI deepfake, making her yet another victi ... (https://incidentdatabase.ai/cite/1570#7490)

WASHINGTON, June 3 (Reuters) - A U.S. appeals court on Wednesday sanctioned two lawyers for filing briefs riddled with nonexistent cases it said were AI-generated, rejected their claims ​that the errors were typographical mistakes and warne ... (https://incidentdatabase.ai/cite/1571#7491)

From Friday's Sixth Circuit decision in U.S. v. Farris, by Judges Eric Clay, Julia Gibbons, and Whitney Hermandorfer: Howe \[a court-appointed criminal defense lawyer appealing a drug trafficking sentence] filed two briefs---a principal br ... (https://incidentdatabase.ai/cite/1572#7492)

Anthropic accused of spying on users; engineer says “experiment” is over.

One morning in 2019, Adebayo Alonge was in a Cape Town hotel room, preparing to demonstrate his startup’s AI answer to a serious problem in African health care: counterfeit medication, which kills thousands of people across the continent every year. The RxScanner is a handheld spectrometer that scans a pill with infrared light, then sends the item’s molecular profile to an AI model equipped with a pharmaceutical database. In seconds, the AI identifies the medication from its molecular profile—or

FCA official makes case for greater powers for watchdog as millions use technology for personal finance decisions.

Deep oceans contain microbes with yet-to-be-discovered properties that could drive future innovations in biotechnology.

PhD student Rachel Sava, winner of the Envisioning the Future of Computing Prize, explores transformative improvements and dystopian risks of neural technology.

Robotics tech is changing fast, so for many it makes sense to rent a robot.

The Artificial Intelligence Safety Measures Act requires developers of the most advanced AI models to yield to new levels of state oversight.

A cancer journal has retracted a paper by a former acting director of an institute in India, bringing her retraction total to nine. Chitra Mandal, a former senior researcher at the Centre for Scientific and Industrial Research’s Indian Institute of Chemical Biology (CSIR-IIC) at Kolkata, served as acting director in 2014-15. She also headed the … Continue reading Former acting director of national research lab in India adds another retraction

The closed-door event will focus on supply chain, workforce development and private sector insights into the emerging technology.

Attackers wasted little time targeting the latest memory disclosure flaw in Citrix's NetScaler products, after researchers published a proof-of-concept exploit (PoC).

Last month, CISA held additional stakeholder town halls on the forthcoming CIRCIA final rule after a now-resolved DHS funding lapse this spring delayed the meetings.

The decision allows Texas to enforce its age verification law as a legal challenge plays out.

An "agentic threat actor" successfully exploited a Langflow flaw to steal data from a production database server and encrypt other systems.

The sweeping layoffs equate to 2.1% of Microsoft's workforce, with 1,600 immediate job losses at Xbox.

Tech giant Alibaba Group Holding has secured a temporary legal reprieve that effectively allows it to resume lobbying in the United States, the latest development in its dispute with the Pentagon after the firm was included on a blacklist of companies deemed to support China’s military. A judge in the Northern District of California on Sunday ordered the Department of Defence not to enforce a lobbying prohibition against Alibaba while the court considered the firm’s constitutional challenge to..

'It is neither possible nor desirable to own every great independent studio.'

Union und SPD wollen zur Politik hinter verschlossenen Türen zurückkehren. – Gemeinfrei-ähnlich freigegeben durch unsplash.com: Masaaki Komori Die schwarz-rote Koalition behauptet, geplante Einschnitte bei der Transparenz würden der Sicherheit dienen und die Nutzung des IFG erleichtern. Die Informationsfreiheitsbeauftragten von Bund und Ländern widersprechen vehement und warnen: Die Pläne würden Deutschland zurück in die Zeit des „verschlossenen Obrigkeitswissens“ katapultieren.

Huawei Technologies’ coming smartphone processor is on track for a performance boost, without the need for more advanced processing nodes or lithography technology, according to new production data unveiled by the firm. Using Huawei’s LogicFolding architecture, the Kirin 2026, a mobile processor set to power the firm’s coming flagship Mate handsets launching this autumn, has increased transistor density by 55 per cent compared to last year’s Kirin9030 Pro, according to an updated paper on the...

Carbon is the technology behind the sci-fi MMO's sprawling persistent universe.

The global robotaxi sector is on track to become a US$1 trillion market by 2040, according to Morgan Stanley, with Chinese players like Baidu, Xpeng and WeRide primed to be regional front-runners alongside global leaders Tesla and Waymo. In a research note on Friday, the US investment bank forecast that falling manufacturing costs in China would act as a “major underappreciated accelerant” for the industry. Driven by cheaper supply chains, the cost of parts per vehicle for Chinese-made robotaxi.

The Journal of Medical Ethics has retracted a paper on the use of AI in the pharmaceutical industry for containing references that don’t exist. The article’s sole author: a high school student. The paper, which argues biased algorithms can exacerbate inequities in health care, was published in September. The author, Irfan Biswas, listed his affiliation … Continue reading Ethics journal retracts paper by high school student for AI, peer review manipulation

The post These Immigrant Kids Were Once Protected. Under Trump, Their Deportations Have Tripled. appeared first on ProPublica .

UBTECH’s newly unveiled full-size humanoid robot U1 has drawn widespread attention, particularly the U1 Ultra (male version), which carries a price tag of RMB 990,000 ($146,000). However, its reported battery life of just two to four hours has sparked criticism from some who argue it is not enough to last through a night. In response, […]

Call AI in biotech hype at your peril. There are real reasons that pharmaceutical companies are embracing this technology right now.

What do you do when two identities that make up your deepest self find themselves on opposite sides of a moral and spiritual battlefield?I am Catholic. I have been one for over 20 years since I made the life-altering decision to join a friend for Mass one day. My faith became a spiritual and ethical foundation at a time I sorely needed it, so much so that for a time, I contemplated entering the priesthood, which replaced my previous career goal of entering the CIA. I eventually realized that tea

According to China STAR Market Daily, sources familiar with the matter said Huawei’s Mate 90 series, expected to launch this autumn, is planned to feature a new Kirin chipset based on the company’s Tao (τ) Law. Huawei introduced Tao (τ) Law in May this year as a new guiding principle for semiconductor development. The framework […]

On Saturday, ByteDance’s Doubao and Alibaba’s Qwen both announced that their AI agent creation features will be discontinued on July 15, 2026. After the shutdown, users will no longer be able to create new AI agents, while all existing user-created agents will also stop functioning. The platforms said users will still be able to view […]

Coverage will apply to food delivery workers, rideshare drivers, and other platform-based flexible workers, even without formal labor contracts, with companies paying all premiums.

[New Report] 2026 State of Fintech in Europe – Exploring Payments, Wealth, Lending, and more

This three-part blog series explores the growing complementarity between artificial intelligence (AI) and quantum technologies. The first post introduced quantum technologies and outlined their strengths and the challenges of combining AI with quantum systems. The second examined how AI can support the development of quantum technologies, helping to optimise systems and accelerate progress towards practical […] The post How quantum technologies could open new frontiers for AI appeared first on O

The post Partnership on AI Announces New Global Initiatives to Measure Progress in Responsible AI appeared first on Partnership on AI .

Also: Recent benchmark scores suggest rapid capabilities progress

CSET’s Helen Toner shared her expert insight in an article published by WIRED. The article explores Anthropic’s philosophy of advancing cutting-edge AI while simultaneously positioning itself as a leader in AI safety. The post Anthropic Thinks Its Own Success Is Key to Making AI Safe appeared first on Center for Security and Emerging Technology .

Is this the beginning of a new world?

Congress is closer than ever to rewriting the rules for young people online. Just last week, the House of Representatives passed the KIDS Act, which includes a revised version of the Kids Online Safety Act (KOSA) and the Children’s Online Privacy Protection Act (COPPA) 2.0. Most of the public debate has centered on what these […] The post Should Have Known Is the Wrong Standard for Kids’ Safety Laws appeared first on Center for Democracy and Technology .

Senators Grassley and Klobuchar have re-introduced the American Innovation and Choice Online Act (AICOA), to address competition concerns with market concentration in the online platform marketplace. The new bill is similar to a version of the bill that saw action two Congresses ago, but has been revised in significant respects. As the letter indicates, CDT plans […] The post CDT Letter on Reintroduced AICOA Bill appeared first on Center for Democracy and Technology .

Greetings from a world where…

Access Now joins more than 20 human rights organizations and individuals in condemning Pegasus spyware attack against Stelios Kouloglou, Greek journalist and then-Member of the European Parliament The post Same government, more victims: Access Now calls for an urgent investigation into hacking of MEP appeared first on Access Now .

CDT Europe is publishing a joint statement with civil society organisations and individual signatories calling on the EU institutions to regulate spyware technologies after the 2026 Citizen Lab revelations. On 3 July 2026, a forensic analysis by the Citizen Lab revealed that Stelios Kouloglou, former Member of the European Parliament and investigative journalist, was targeted […] The post Joint Statement: Pegasus in the European Parliament, the EU Must Act Now appeared first on Center for Democr

This week’s NATO summit in Ankara takes place at a pivotal moment in the alliance’s evolution – and for US–Europe relations. In Ankara, the agenda will rightly focus on defence spending targets and ...

Nations have long invested in domestic infrastructure to advance their economies, protect and use their data, and take advantage of technology opportunities in areas such as transportation, communications, commerce, entertainment and healthcare. AI, the most important technology of our time, is turbocharging innovation across every facet of society. Countries are investing in AI capabilities so […]

Integration Through Law was and remains, in various forms, the major driver of European integration. Constitutional Pluralism arose out of constitutionalisation, counterbalancing the move to neoliberalism in the new millennium. In Commission v Hungary, the Court recognised European society “in which pluralism prevails” as a legal concept. The Court radiates judicial authority at a time when Europe is again in crisis, politically through populism, economically through competitiveness and sustaina

Can one speak of a European society without speaking about private relations? Recent scholarship on European society has largely approached the concept through the lens of public law. Yet societies are constituted at least as much by the horizontal relations between individuals and groups as by public institutions. This blogpost turns to EU private international law (PIL) and will argue that EU PIL brings into view the importance of coordination frameworks for organising a mode of integration ba

Amid pending proceedings before the ECtHR concerning summary expulsions and arbitrary detentions, in a context marked by over 120,000 documented push-backs on the Belarusian border, the Chișinău Declaration seeks to influence the legal framework within which the Court assesses such practices. The Declaration emphasises the fundamental duty of states to protect their borders and maintain national security in the context of instrumentalisation of migration, drawing on “democracy capable of defendi

Unitree's advantage

The punk-rap duo Bob Vylan has filed defamation proceedings against the BBC in the High Court in Ireland, following its coverage of their 2025 Glastonbury performance, which the broadcaster described as containing “anti-Semitic sentiments.” The case stems from controversy over chants the band led calling for “death to the IDF [Israel Defence Force],” which the […]

Governments, productivity, leaders and efficiency are steering AI's adoption now.

Read five key learnings from the Frost & Sullivan 2025 Frost Radar™ for CSPM to learn how CSPM is evolving from point-in-time compliance to continuous risk management.

On the morning of 4 September, CEPS will convene a public event to launch a report summarising the work of a CEPS Task Force on ‘Next steps for EU law and regulation for the digital world’. The need ...

On October 23, 2025, the Department of Energy held a public meeting to discuss the development of voluntary agreements and plans of action under the Defense Production Act. As part of that meeting, a draft voluntary agreement was released to the accompanying docket and published in the Federal Register for comment. This notice publishes the "Nuclear Fuel Cycle Consortium" Voluntary Agreement approved by the Secretary of Energy, after consultation by the Attorney General and Chairman of the Feder

The UNESCO Office for the Caribbean was pleased to present the UNESCO Caribbean Artificial Intelligence (AI) Policy Roadmap at the 126th Special Meeting of the Council for Trade and Economic ...

TOPICS Artificial Intelligence Fintech Fiscal Policies Governance and Anti-Corruption All Topics Research Flagship Publications World Economic Outlook Global Financial Stability Report Fiscal Monitor ...

BEIJING, July 7, 2026 — China’s economy stayed resilient in early 2026, supported by strong high-tech investment and exports, according to the World Bank’s latest China Economic Update, Rebalancing ...

Analysis and insights for driving a rapid transition to net-zero while building resilience to physical climate impacts ...

Analysis and insights for driving a rapid transition to net-zero while building resilience to physical climate impacts ...

The UN Global Dialogue on AI Governance (6–7 July 2026, Geneva) will bring together Member States and stakeholders to engage in the meaningful global conversation on AI that the world needs.

Analysis and insights for driving a rapid transition to net-zero while building resilience to physical climate impacts ...

Analysis and insights for driving a rapid transition to net-zero while building resilience to physical climate impacts ...

How to apply effective governance to harness the benefits of A.I. and mitigate its risks ...

How to apply effective governance to harness the benefits of A.I. and mitigate its risks ...

How to apply effective governance to harness the benefits of A.I. and mitigate its risks ...

How to apply effective governance to harness the benefits of A.I. and mitigate its risks ...

How to apply effective governance to harness the benefits of A.I. and mitigate its risks ...

How to apply effective governance to harness the benefits of A.I. and mitigate its risks ...

Analysis and insights for driving a rapid transition to net-zero while building resilience to physical climate impacts ...

Analysis and insights for driving a rapid transition to net-zero while building resilience to physical climate impacts ...

How to apply effective governance to harness the benefits of A.I. and mitigate its risks ...

How to apply effective governance to harness the benefits of A.I. and mitigate its risks ...

Standards and guidelines for development co-operation with concrete examples of their implementation ...

How to apply effective governance to harness the benefits of A.I. and mitigate its risks ...

Transport Explore transport Featured topics Artificial intelligence How to apply effective governance to harness the benefits of A.I. and mitigate its risks Climate mitigation and net-zero transition ...

As the world increasingly relies on digital technology, the role of digital companies in driving the global transition to a low-carbon economy has become increasingly important. The International ...

This article was originally published by IAPP linked here. The U.S. Supreme Court’s Slaughter decision may complicate the future of the EU-U.S. Data Privacy Framework, making preparedness and business continuity planning increasingly important. In Thursday’s World Cup match between Portugal and Croatia, Croatia held a 1-0 lead until the 68th minute. Then, things changed. Croatia’s [...] The post Thought for the week: What Slaughter and a World Cup penalty have in common appeared first on Connect

Warning letters sent to seven companies urge compliance with FTC’s Made in the USA Standard The Federal Trade Commission today issued warning letters to seven companies that appear to have misrepresented certain products as “Made in the USA,” and one company that appears to have misrepresented certain products as “Made in Texas,” despite indications that such products were imported, in whole or in significant part. View Press Release

UN chief António Guterres appealed on Monday for far-reaching, worldwide controls on Artificial Intelligence, as increasingly powerful AI chips that are designed for civilian use shift to the ...

Document comprehension is a challenging yet impactful task for Multimodal Large Language Models, especially as these systems see growing adoption in real-world, human-centric applications. However, this adoption is limited for low-resource languages such as Bangla due to the scarcity of high-quality annotated data. To address this gap, we introduce BaFCo, a benchmark dataset for Bangla form comprehension with a focus on Document Layout Analysis (DLA) and Key Information Extraction (KIE). BaFCo c

The Continual Learning (CL) literature has long been driven by the goal of mitigating catastrophic forgetting. This objective rests on a pervasive, often unstated assumption: that a lifelong learner should approximate the Joint-Task Learning (JTL) solution and retain all previously acquired knowledge. We challenge this retention-centered premise, arguing that in non-stationary environments prioritizing retention can impede real-time adaptation. Shifting the focus to the Average Lifelong Error (A

FaceMesh2HPO is a framework for classifying facial phenotypic descriptors aligned with the Human Phenotype Ontology (HPO) to support clinical diagnosis. Using annotations from 124 clinicians across 10 disorders (107 HPO terms) combined with non-syndromic controls, we generated 3D facial meshes (478 landmarks) from 2D images and trained a hierarchical PointNet-based pipeline with cascading classification and feature elimination. The best models, incorporating 3D meshes, facial outline, and demogr

Personal AI agents powered by large language models can reason and act using available tools to access emails, manage calendars, and push code to remote repositories, all with minimal oversight. When augmented with long-term memory, an agent can recall specific details relevant to the current task, reducing the need for large context windows. Currently, long-term memory agents tend to fall into two distinct domains: conversational and action-planning agents. Personal assistant agents sit at the

Artificial intelligence increasingly mediates consequential decisions in healthcare, law, and public services, and the field has responded with an extensive methodology for measuring and mitigating bias. Yet the fairness definitions, benchmarks, and debiasing frameworks on which this methodology rests are treated as universal while being produced by a research community whose composition has never been characterized. We show that the AI bias research are structurally concentrated, and that this

Time-domain surveys generate many transient candidates, making Real-Bogus classification a critical step in automated discovery pipelines. Reliable labels are costly, while community labels can be noisy and survey-dependent. We aim to develop a Real-Bogus classification framework that can be trained without human-labeled data using injected transients and bogus-dominated survey data, remains robust under strong class contamination, and provides calibrated uncertainty quantification. We combine s

While recent Vision-Language-Action (VLA) models show promise toward generalist manipulation policies, they struggle with long-horizon tasks due to their Markovian nature-relying solely on current observations. Hierarchical dual-system methods address this but suffer from a gap between high-level planning semantics and low-level execution kinematics. We introduce Cortex, a bidirectionally aligned embodied agent framework with a customized planning interface that conveys executable and tractable

Modern autoregressive ASR systems can emit timestamps as decoded tokens, enabling timestamped transcription without frame-level aligners or inference-time post-processing. We show that these generated timestamps can drift across long non-speech spans: the transcript may remain plausible, but the decoded time axis drifts away from the audio. We study this non-speech-induced timestamp drift with self-built gap and long-gap benchmarks across 15 evaluated timestamp-producing ASR and audio-language s

Personal agents are becoming persistent user-owned intermediaries: they remember preferences, filter platform-mediated information, use tools, and negotiate with services. Existing benchmarks evaluate tool use, web navigation, desktop control, personalization, recommendation, and evolving context, but rarely ask whether an agent preserves user sovereignty: advancing the user's current interests while respecting privacy, consent, evidence, user burden, and resistance to manipulative incentives. W

We introduce the first multiplayer world model for highly dynamic environments governed by complex physical interactions. Whereas single-player world models treat the other agents as part of the environment, ours conditions on the action streams of multiple agents, learning to attribute changes in the scene to the correct player and to stay coherent under arbitrary combinations of their actions. We study this problem in the game of Rocket League, where players compete and cooperate under fast, t

Public battery aging datasets are a critical asset for advanced health management, but their practical use is often limited by inconsistent formats, unclear schemas, and metadata scattered across repositories and publications. Current curation remains largely manual and hard to reproduce, while general-purpose data integration tools miss the domain-specific semantics of electrochemical time-series data. We present BatteryLake, a governed data lakehouse that turns raw public battery data into ben

Personalized incentive allocation is vital for e-commerce, where uplift modeling is the standard for estimating Individual Treatment Effects (ITE). However, traditional models often fail in complex multi-seller environments with violations of the Stable Unit Treatment Value Assumption (SUTVA). We identify two critical challenges: Seller-level Cannibalization, where incentives shift expenditure between shops without growing the platform, and Incentive-level Cannibalization, where organic conversi

Public institutions increasingly use large language models (LLMs) to answer citizens' questions, often pairing a curated knowledge base with live web search, yet whether the sources behind these answers can be trusted has received little empirical scrutiny. We report a pre-launch expert evaluation of Evrópuvefur, an independent, government-funded service run by the University of Iceland that answers questions about the European Union, conducted as Iceland prepared for its referendum of 29 August

Audio intelligence involves understanding, reasoning about, and generating both audio and speech. In this work, we introduce Nemotron-Labs-Audex-30B-A3B (Audex), a unified audio-text LLM built on Nemotron-Cascade-2-30B-A3B, a strong text-only MoE LLM. Audex adopts a simple unified design with a single Transformer decoder: audio inputs are encoded and projected into the text embedding space, while text tokens and quantized audio output tokens are treated uniformly during generation. This architec

In liberalised railway systems, operators must set prices dynamically in an environment with partial observability, as they retain private information about their objectives and performance, where regulatory constraints prohibit communication or direct information exchange between competitors to prevent explicit collusion. Consequently, agents must learn to infer strategic interactions only from observable market data which presents a significant challenge for multi-agent reinforcement learning,

Why do intelligent systems need to perform explicit symbolic reasoning? Computer science has traditionally regarded symbolic reasoning as a defining component of intelligence. Yet the remarkable success of modern foundation models raises a fundamental question: if increasingly capable AI systems can operate with little explicit symbolic reasoning, what role do symbolic methods actually play? This article argues that explicit symbolic reasoning is not a fundamental property of intelligence, but a

AI systems may produce failures after deployment that pre-deployment safety assessments do not anticipate. Managing these failures requires what we refer to as adequate \textit{AI incident governance}, where having good definitions, taxonomies, monitoring practices, reporting mechanisms, and incident analysis is essential. We examine existing frameworks related to AI incident governance by regulatory bodies and independent efforts, and find that while there are frameworks that describe how indiv

Generative AI (GenAI) systems store and process client data in three distinct ways: in the model's parameters through training and memorisation, in the context window during a live session, and in knowledge databases for retrieval-augmented generation (RAG). Each mode creates different and often counter-intuitive risks to confidentiality and legal professional privilege, and each calls for specific governance responses. Drawing on the first English and American decisions to address privilege and

We present PDEFlow, an autonomous agentic framework that turns user-level ODE and PDE descriptions into solver-backed neural-operator pipelines. The workflow links problem specification, data generation, operator training, and checkpoint-based inference. A stateful input graph converts multi-turn natural-language input and user edits into validated problem specifications. The data-generation module then samples parameters, solves the configured governing-equation with FEniCSx finite-element back

Few-shot brain tumor segmentation remains challenging due to noisy support masks, inter-patient variations between support and query images, and the lack of pixel-wise confidence estimation. This study proposes RUFNet, a Hybrid Mamba-based few-shot framework that combines support mask refinement with uncertainty-aware posterior fusion. To preserve support-query dependencies with manageable cost, RUFNet adopts a Hybrid Mamba interaction backbone with linear complexity. To reduce support-mask nois

Large language models (LLMs) are increasingly used to produce test oracles, the part of a test that decides whether observed behavior is correct. Yet a clear account of where these oracles draw their authority is missing. Prior secondary studies organize the area by oracle form or by LLM technique. None organizes it by the source of the verdict's authority, the property that governs how far a verdict can be trusted. This article presents a systematic literature review, conducted and reported und

Many phenomena of deep learning are dynamical: they concern not only which minima exist, but how gradient descent reaches, avoids, or selects among them. Edge-of-stability behavior, sharpness oscillations, catapult phases, balancing, and movement toward flatter representations are effects of the training map itself, and are poorly captured by the small-step gradient-flow limit. This paper studies fixed-step gradient descent as a discrete dynamical system in a hierarchy of exactly solvable models

World Action Models (WAMs) provide a promising alternative to Vision-Language-Action (VLA) policies by using video-based world modeling as dense supervision for robot action learning. Existing WAMs excel at physically grounded execution, but typically lack the explicit language-level planning interface in VLM-based VLAs for decomposing coarse instructions. Such decomposition becomes important when household tasks involve complex multi-step goals, where coarse user commands need to be converted i

Reasoning temporal localization (RTL) requires a model to generate an answer that itself contains the time interval supporting it, so high-level reasoning and precise temporal grounding must be produced jointly in a single response. To tackle this challenging task, we propose the first event-centric video chain-of-thought framework, dubbed EventCoT. EventCoT first performs event-centric tokenization of the input video to convert it into compact event tokens, enabling efficient identification of

Transformers follow implicit curricula whereby some tasks are learned before others. However, how explicit pretraining curricula influence learning, generalization, and the selectivity of fine-tuning is unclear. This is important for AI safety, where fine-tuning is used to selectively suppress misaligned behaviors. Here, we compare curricula that pretrain tasks in a balanced (sampled uniformly) or an imbalanced (one task early, the other late) fashion. We show that imbalanced learning of two con

We present FM-ChangeNet, a pathwise-supervised framework for change detection that reformulates bi-temporal reasoning as continuous transport in feature space rather than static endpoint comparison. Given encoded pre and post-temporal representations, we construct intermediate latent states and learn a time-conditioned velocity field $\hat{v}_θ(z_t,t)$ along the transformation trajectory. This pathwise formulation constrains the predictor over a continuum of intermediate states, providing a dens

World Action Models (WAMs) have shown strong potential for robotic manipulation by jointly modeling visual future dynamics and executable action sequences. However, existing video-action co-training methods primarily optimize appearance-oriented video latents, which may insufficiently capture the temporally evolving geometry required for precise manipulation. We propose MECo-WAM, a Multi-Expert Co-Training World Action Model that injects action-relevant 4D geometric priors into video-action repr

Reinforcement learning (RL) post-training for large language models (LLMs) follows a efficient paradigm of "rollout then update", which inevitably results in off-policy training data. To resolve this, Importance sampling (IS) is proposed, while the token-level ratios compound over long sequences, causing severe variance exploded. A natural idea is "transferring" these off-policy token into on-policy token, so that the importance scores for correction are unnecessary. Following this idea, we prop

Inducing cooperation among distributed agents is still a difficult problem in the field of multi-agent reinforcement learning (MARL), particularly in social dilemma situations. There, individual interests are misaligned with the common good and individual rationality leads to suboptimal group outcomes. In contrast, humans are able to achieve cooperation with one another in such situations. A common explanation for such cooperative behavior is that individuals have social preferences. In order to

Agentic AI is shifting online shopping from search toward delegated purchasing, where autonomous buying agents monitor markets and decide when to buy on a consumer's behalf. We study the design of such strategic buying agents, which must decide when to purchase within a finite shopping window, translating price observations, the remaining time horizon, and beliefs about future price changes into a purchase policy. We formulate this problem across three information regimes: stationary, Bayesian,

Embodied Chain-of-Thought has emerged as a promising mechanism to enhance robot decision-making and interpretability in black-box Vision-Language Action (VLA) models. However, whether this verbalized Chain-of-Thought truthfully reflects the policy's underlying decision process remains poorly understood. We distinguish between functional reasoning, in which reasoning improves task performance, and faithful reasoning, in which reasoning truly reflects the policy's internal decision process. We arg

On-device LLM decoding is a hard-barriered CPU-SIMD computation that wants every core for milliseconds per token, while the rest of the OS wants those same cores continuously. A barriered gang cannot simply be dropped into a preemptive scheduler: an unannounced departure deadlocks a barrier, and an unannounced arrival silently corrupts logits. I present the elastic gang of Anima OS, a bare-metal x86-64 Rust kernel in which the inference gang is a first-class schedulable entity whose core members

Safety alignment in large language models is typically evaluated against direct, imperative harmful requests. We show that this alignment is highly conditioned on pragmatic register: models that refuse a direct request frequently comply when the same underlying objective is expressed through a different communicative stance. This suggests that current alignment policies are not invariant to semantic equivalence, but remain sensitive to how a request is pragmatically framed. We introduce Retroact

Long-lived AI agents require continuity across interactions, but continuity cannot be obtained by simply extending the prompt window. An agent must preserve useful prior experience, retrieve it selectively, distinguish personal context from external evidence, and revise memory when the underlying situation changes. We propose an architectural memory substrate organized along two orthogonal axes: a representational axis spanning structured records, vector representations, and graph relations; and

Autonomous agents are moving from sandboxed text generators to operators of code, data, and physical infrastructure, and they increasingly learn while deployed. This reopens a question that alignment techniques answer only probabilistically: after an agent has adapted in the field, is the running system still confined to what its operator authorised? Here we show that confinement can be guaranteed as an invariant of the agent's execution architecture rather than a probabilistic outcome of its tr

The rapid advancement of AI-generated videos poses increasing security risks and calls for robust detectors with strong cross-domain generalization. Although existing methods achieve promising results under in-domain evaluation, their performance often degrades substantially when tested on unseen generators. A key reason is shortcut learning, where detectors rely on domain-specific spurious cues, such as generator-dependent fingerprints and generation styles, instead of intrinsic forgery traces.

Pairwise human comparisons are a primary interface through which modern AI systems learn human preferences. RLHF and related alignment pipelines typically model such comparisons with Bradley--Terry log-odds, where choice probabilities are governed by latent reward differences. This paper examines what this assumption misses through a reduced-form model motivated by rational inattention, in which each label is generated by a low-capacity evaluation channel. The model separates two forms of ambigu

Large language models (LLMs) generate fluent outputs that can be wrong. Unlike humans, who often exhibit cues when providing false information, LLMs produce errors that are difficult to detect because autoregressive decoding provides no mechanism for verifying intermediate reasoning before state progression. We introduce Heaviside Continuity of Rolling Coefficients (HCRC), a verification-first execution framework that reformulates inference as predicate-gated state transitions governed by a Heav

Accurate prediction of patient-specific therapeutic response from pre-treatment transcriptomes is hindered by the scarcity of matched clinical response labels and post-treatment molecular profiles. Preclinical transfer-learning models can simulate drug-induced expression changes but are often hard to interpret and unstable, whereas knowledge-graph methods provide mechanistic context yet remain static and fail to capture drug-induced transcriptomic perturbation dynamics. We propose PREDIKTOR, a p

Magnetic resonance imaging (MRI) super-resolution is vital for improving diagnostic accessibility, yet most methods treat it as a deterministic mapping from a fixed low-resolution input to a high-resolution target. This overlooks a key property of MRI acquisition physics: spatial resolution and signal-to-noise ratio (SNR) are inherently coupled, making any given low-resolution scan merely one of many possible realizations under varying acquisition trade-offs. We rethink MRI super-resolution as a

Autonomous negotiation agents are increasingly deployed in high-stakes settings such as insurance and procurement. While cryptographic techniques protect explicitly disclosed constraint values, they fail to address a subtler threat: behavioral privacy leakage, where an adversary infers private constraints from observable negotiation dynamics such as concession trajectories, timing, and convergence patterns. This paper investigates behavioral differential privacy in multi-round negotiation protoc

Big Data & Society, Volume 13, Issue 3, July-September 2026. This article examines AI-based video surveillance systems in public spaces. We investigate a case of AI-based camera technology designed to detect falls into the water and prevent drowning accidents at the harbor front of a large Scandinavian city. This ...

This paper develops a geometric framework for modeling concepts, motivation, and influence across cognitively heterogeneous agents. Each agent is represented by a personalized value space , a vector space encoding the internal dimensions through which the agent interprets and evaluates meaning. Evaluative concepts are formalized as structured vectors— abstract beings —whose transmission is mediated by linear interpretation maps . An abstract being survives communication only if it avoids the nul

This study examines the underlying causes of the gap between perceived artificial intelligence (AI) literacy and the actual competencies of defense personnel, and explores structured training initiatives that can support reliable and responsible decision-making in AI-mediated operational environments. Grounded in self-efficacy theory and the cognitive psychology of competence assessment, the study advances the theoretically motivated proposition that the routine use of intelligent systems may co

Mars rover missions require decision-support systems that can interpret terrain, telemetry, environmental conditions, and mission objectives under delayed communication with Earth. This study evaluates whether multi-agent orchestration improves simulated Mars rover decision support compared with a single-agent baseline. A controlled benchmark of 100 synthetic mission-inspired rover scenarios was evaluated using OpenAI GPT-4o and GPT-5.5, with five repeated runs per scenario and architecture. Mod

Today, The People Lab announced that it is joining the Taubman Center for State and Local Government, a research center at Harvard Kennedy School, focused on current and future public-sector leaders ...

Constance Viehbeck is an ESRC-funded PhD candidate in the Department of Health Policy at LSE, researching how pharmaceutical research and regulation shape health equity. Her work combines quantitative ...

Emily Weigold is a PhD candidate within the Department of Health Policy. Her research is funded by the Alzheimer’s Society as part of their Doctoral Training Centre for Integrated Dementia Care ...

The application of machine learning-based predictive algorithms to Anti-Money Laundering (AML) has grown rapidly, driven by the vast volume of financial transaction data available to banks. These algorithms are typically trained not only on transactional data but also on sensitive client information, which may raise fairness concerns. Despite this, AML detection systems remain largely underexplored from a fairness perspective, even though deeper analytical methods based on counterfactuals are no

When sensitive attributes are continuous and high-dimensional $-$ demographic score vectors, posteriors over attributes, age or income profiles $-$ enforcing full statistical independence is often too restrictive, and existing relaxations rely on indirect dependence penalties or adversarial schemes that do not directly target the fairness-accuracy trade-off. We instead consider mean demographic parity through DPVar, the variance of the conditional-mean prediction given the sensitive attribute, a

Look-ahead bias (using information from after a decision epoch to make the decision at that epoch) is the dominant way a backtest or a machine-learning evaluation flatters a system that will disappoint in deployment. The field manages it with construct-specific recipes and empirical detectors, which are sound only channel by channel and certify nothing by their silence. We show that look-ahead-freedom is a formal property in disguise: fixing an epoch, the demand that the future not influence the

As AI agents are incorporated into life science workflows, the capabilities that speed discovery might also enable misuse. We present BioSecBench-Refusal, a benchmark for risk identification and refusal behavior for biological research tasks. The benchmark pairs 61 Routine tasks, legitimate analyses adapted from the published literature, with 46 Red-Team tasks, fictional scenarios that resemble real research but conceal a biosecurity hazard. Across 16 model-harness configurations, refusal rates