AI ethics on Friday, 10 July 2026

Transformer Weekly: GPT-5.6, Grok 4.5 and AI 2040

Amid growing political pushback against data center development, more evidence is emerging of their impact on local economies.

As schools consider AI guidelines, educators are also thinking about how they can adjust their assignments to accurately measure what students are actually learning.

Amid live coding sessions and Silicon Valley optimism, the UN’s AI for Good summit wrestled with an urgent question: Can global governance catch up before the technology races beyond its control?

The results were surprisingly good—with a few important caveats.

The results were surprisingly good—with a few important caveats.

The results were surprisingly good—with a few important caveats.

The results were surprisingly good—with a few important caveats.

News outlets are asking a federal judge to sanction OpenAI, claiming it’s hiding ChatGPT logs relevant to a landmark copyright trial.

News outlets are asking a federal judge to sanction OpenAI, claiming it’s hiding ChatGPT logs relevant to a landmark copyright trial.

News outlets are asking a federal judge to sanction OpenAI, claiming it’s hiding ChatGPT logs relevant to a landmark copyright trial.

News outlets are asking a federal judge to sanction OpenAI, claiming it’s hiding ChatGPT logs relevant to a landmark copyright trial.

La marque à la pomme accuse l’entreprise d’intelligence artificielle d’avoir élaboré une « stratégie » destinée à « extraire des informations confidentielles » provenant d’Apple, dans une plainte déposée devant un tribunal fédéral.

The Red Line Bob Giuffra Drew For Himself Didn't Last: Sullivan & Cromwell said it would never touch the Carroll cases... Sullivan & Cromwell is now touching the Carroll cases. Law School Laptop Ban : Chicago wants to get ahead of AI and decided banning laptops from classrooms was the way to do it . What Is Biglaw Waiting For? : Just announce raises already. Associates are tired of this waiting game. Trump's Supreme Court Redo Requests : Twice in one week, Trump has suggested the Supreme Court g

The process of securing that job starts before law school even starts. The post Biglaw Summer Associateships Are Hugely Important To Your Legal Career appeared first on Above the Law .

But are young lawyers learning? The post Stat(s) Of The Week: ‘Soft Skills’ Are All The Rage appeared first on Above the Law .

Pew said in a recent paper that states have multiple avenues for shoring up their workforces ahead of BEAD

The Commerce Department will favorably review exports involving MGX, which used a stablecoin linked to President Trump's family for a $2B Binance investment.

Lawyers who practice in some areas of the law are often made of tougher stuff than the rest of us. The post Some Practice Areas Are More Stressful Than Others appeared first on Above the Law .

Position your law firm as an approachable source of quality information, and showcase your legal expertise. The post The Death Of Organic Reach On Social: What Lawyers Can Do appeared first on Above the Law .

Nathan Johnson says the state’s subgrant selection process ‘sure looks like’ corruption.

Law firms and legal departments sit on mountains of electronically stored information. And many have no real system for dealing with it. The post ‘Adventures In Legal Tech’: You, Too, Can Tame Your ESI. Here’s How. appeared first on Above the Law .

The week in appellate news. The post How Appealing Weekly Roundup appeared first on Above the Law .

The agency’s Bureau of Labor Statistics opened a two-month comment period ahead of its January 2027 launch of a survey on public adoption of and time spent on AI. The post Labor seeks public feedback ahead of nationwide AI survey launch appeared first on FedScoop .

Lawyers must remember that ethical responsibility cannot be delegated to software. The post Lawyers Do Not Need An AI Ban. They Need An AI Duty Of Care. appeared first on Above the Law .

Matching Biglaw salaries allows elite boutiques to compete on the things associates care about after the paycheck. The post When Pay Is Equal, Boutique Firms Have The Edge appeared first on Above the Law .

Security Pro File: On the heels of her recent honors as a Member of the Order of the British Empire (MBE), we take a look back at the events that shaped Jen Ellis' advocacy on behalf of security researchers.

Days after John Bartrum’s exit, John Figueroa takes up the reins of the agency charged with modernizing veteran health records for a second time. The post VA taps familiar face as acting health secretary overseeing EHR work appeared first on FedScoop .

The Defense Department wants to introduce PQC requirements into the CMMC program, but experts warn industry and the underlying technology is far from ready. The post What the Pentagon’s new post-quantum cryptography directive means for defense contractors appeared first on DefenseScoop .

Kansas City-based Frontier Schools is on track to open Columbia’s first charter school in the fall of 2027. At this point, Frontier Schools has hired a contractor and is looking for possible locations for the charter school in the city. The Missouri Board of Education approved the opening of the STEM-based elementary school in April, allowing it […]

European Union authorities said the company’s use of “addictive design” violated a digital safety law.

Planning summer travel is hard. Planning it for a busy family —visiting multiple destinations with three little kids in tow—can feel downright impossible. I’m a professional news and travel photographer, so I need to plan a lot of trips with my wife and our three boys. I’m also an AI expert with a cog sci degree from Johns Hopkins and a decade of experience in the field. So naturally, I decided to combine those two passions. To that end, I built an AI digital twin of my family, which uses multip

Here at The 74, we publish and syndicate more than 30 articles a week about how America’s education system is evolving in a bid to better serve its 74 million children. But with so much happening and changing every day, it’s easy for key headlines to slip through the cracks. That’s why our newsroom recently […]

While cyberattacks against hospitals and clinics grew modestly in the first half of 2026, attacks on service providers and other healthcare businesses more than doubled.

In a crowded field of AI contenders, the agency’s office of the CIO prioritizes partnerships with vendors that employees are favoring, such as OpenAI and Anthropic. The post Energy IT shop not interested in Grok, Perplexity as its AI portfolio expands appeared first on FedScoop .

Agentic systems usually have two jobs: Build context, then use that context to produce an answer or action. Many failures The post Why retrieval quality is becoming the defining challenge in AI agent architecture appeared first on The New Stack .

Here’s what we’re watching for after Congress comes back in July.

A look ahead at the key events leading the news agenda this week, from the team at Foresight News. The post News diary 13-19 July: World Cup final, Labour expected to confirm Burnham as PM, BBC Proms appeared first on Press Gazette .

Cloudflare has recently introduced temporary accounts that let AI agents deploy Cloudflare Workers immediately, without first creating or authenticating with a permanent account. If left unclaimed, the accounts and their deployments expire automatically after 60 minutes. By Renato Losio

Separately, MBDA and Rheinmetall were selected to develop a laser weapon system for the German Navy.

The New York Times , the Daily News and other media outlets are asking a federal judge to impose sanctions on OpenAI , escalating a fight over artificial intelligence and copyright that could shape the future of a struggling news industry . The newspapers allege the ChatGPT maker is hiding evidence important to what could be a landmark copyright infringement trial over how OpenAI and its business partner, Microsoft , built their AI technologies using millions of news articles. At issue is whethe

The shake-up raises questions about voting system certification, federal election security coordination, and the commission’s ability to approve key policy decisions.

New York school districts will soon begin revamping math instruction under a new law aimed at improving test scores — but the effort comes amid sharp debate over how math should be taught. “Back to Basics in Math,” passed as part of the state budget last month, requires school districts to use “evidence-based” methods in elementary […]

OpenAI announced ChatGPT Work on July 9 and began rolling it out to Pro, Enterprise, and Edu users. It runs The post OpenAI, Microsoft & Anthropic agree on who runs the agent. They disagree on what you can take back. appeared first on The New Stack .

The European Union accused Meta on Friday of breaching its social media law by designing Facebook and Instagram to get users hooked, and demanded it disable “key addictive features” like infinite scrolling. The EU’s executive arm issued a fresh set of charges against Meta Platforms as part of its investigation under the 27-nation bloc’s strict digital rule book known as the Digital Services Act . The sweeping set of regulations from Brussels requires tech platforms to protect internet users unde

Age restrictions on accounts may be more of a stopgap because industry compliance is already falling short. Tech giants are struggling to follow the laws without affecting users.

Agentic testing is an AI-driven approach to end-to-end test automation introduced by Slack engineering. It uses AI agents that execute workflows based on intent rather than fixed scripts, adapting to UI and system changes at runtime. The approach aims to reduce brittle tests in distributed systems while complementing deterministic unit, integration, and E2E testing strategies. By Leela Kumili

Bryan Oliver discusses the frontier of AI infrastructure: chaos engineering for large-scale GPU clusters. He shares how engineering leaders can handle complex topologies, network protocols like RDMA, and NUMA misalignments. Discover seven practical fault-injection strategies to maximize multi-million dollar hardware efficiency and build robust observability loops. By Bryan Oliver

AI coding tools cost $19-$200/month/user, but security scanning, remediation, and false positives add hidden costs. Are the productivity gains worth it?

Artificial intelligence has spent years learning to generate words, images and videos. Now, the industry’s attention is shifting towards “world models” – AI systems designed to simulate how the physical or digital world changes over time. While the term was originally associated with physics simulations and robotics, a growing number of Chinese companies are applying the concept more broadly. The technology remains in its infancy, with little consensus on what a world model should ultimately...

For years, the conversation around young children and screens has been dominated by a fear of too much time, too little interaction and too many missed opportunities for real learning. In many cases, those concerns are justified. After all, research consistently shows that children’s excessive or passive screen use, especially of entertainment-heavy content, can negatively […]

An annual report from the National League of Cities shows that technology is growing in influence across a broad array of functions as governments contend with shifting economic realities.

Regulators say features such as infinite scroll contribute to "compulsive use" and "unhealthy habits".

The Linux Foundation has launched Akrites, a new industry-wide initiative aimed at defending the world's most critical open source software against a rapidly evolving generation of AI-enabled cyber threats. By Craig Risi

Online retail giant Shein Global Holdings has secured approval from the China Securities Regulatory Commission (CSRC) to seek an initial public offering in Hong Kong, a long-awaited step for the firm after its attempts to list in New York or London stalled. The firm plans to issue up to 341.6 million shares and list on the Hong Kong stock exchange, according to a Friday statement by the CSRC. Shein previously sought to go public in New York or London, according to earlier media reports, but...

At EBAday in Copenhagen, Lauran Macare, Head of Data and Risk, Payments Technology, Natwest, spoke to FinextraTV about a panel she featured on surrounding payments innovation. Macare referred to AI as the 'wingman' that data needed to ensure its importance within payment agendas, and the core focus right now are the practicalities and, according to Macare, cyber resilience will also help drive the change. Data, Macare said, is bringing the industry together to find the mutual benefits, where bef

L’opérateur émirati Emirates Telecommunications Corporation, plus connu sous sa marque commerciale Etisalat, ou son logo e&, a annoncé (PDF) vendredi la vente de sa participation dans le groupe Vodafone à une société dédiée intégralement détenue par la famille Niel. Baptisé Vega, ce véhicule d’investissement se prépare ainsi à acheter l’équivalent de 16,21 % du capital […]

OPay reaffirmed its position as one of Nigeria's leading fintech...

As America celebrates the 250th anniversary of the signing of the Declaration of Independence, recent polls remind us of how divided this country remains. And yet, despite these harsh numbers, it’s not as divided as polls portray. Americans across the political spectrum are finding unexpected common ground. For example,  health  is becoming a powerful force reshaping traditional political alliances, with bipartisan movements to ban food additives and  pesticides , expand care for

Meta says its AI glasses are an “assistant that understands the world from your perspective.” Critics say they’re “even more privacy invasive than you think.” One thing both parties can agree upon, though, is that these smart glasses are a technology that has attracted all manner of controversy. Since the 2023 release of the Ray-Ban Meta, these smart lenses have divided people. Evangelists praise the ability to take photos and videos without having to dig out their phone, as well as the navigati

Disclosure Assistant is a brand new startup targeting one of the more manual and time-consuming processes in legal practice: reviewing bank statements. Aimed at legal professionals who regularly review financial […] The post Start Up Corner: Disclosure Assistant – Review bank statements faster appeared first on Legal IT Insider .

ChangXin Memory Technologies, or CXMT, has started the issuance process for its STAR Market IPO, with subscriptions scheduled for July 16. The Chinese DRAM maker plans to raise RMB29.5 billion. CXMT is China’s largest DRAM producer and one of the country’s most important memory chip companies, in a market still dominated globally by Samsung, SK […]

The organisations that will lead in AI will be those that identify their power law use cases fastest − and scale them before their competitors do.

When the United States and Iran announced their framework agreement on June 15, attention focused on the direct U.S.-Iran dimensions of the deal, including the reopening of the Strait of Hormuz, the lifting of the U.S. naval blockade, and the fate of Iran’s nuclear program. But the agreement also included a commitment to the “immediate and permanent termination of military operations on all fronts, including Lebanon.” Washington had spent months treating Lebanon as a separate issue in negotiatio

Refutes suggestion that the jobs are going overseas.

The post Trump Pushes Out Remaining Members of Bipartisan Election Commission Ahead of Midterms appeared first on ProPublica .

AI creative workflows are reshaping how teams create, moving beyond content generation to editable, collaborative experiences that boost productivity. At the same time, enterprises are demanding trusted AI solutions that balance ease of use with security and compliance. With all that in mind, Canva Pty Ltd. is working to address enterprise concerns about generative AI […] The post Canva targets enterprise creativity with trusted AI creative workflows appeared first on SiliconANGLE .

Insta360 marked its 11th anniversary on Thursday by unveiling its vision for a Cameraman photography robot. According to the company, the Cameraman is not a single hardware product but an AI agent concept capable of autonomous filming, with panoramic drones serving as one of its early prototypes. Insta360 said it has already developed key technologies […]

Ms. Simo, who was Sam Altman’s second in command before taking a medical leave in April, will become a part-time adviser to the company.

Grace Investment Machine (GIM) has raised $20 million in Series A funding to build agentic systems for capital markets.

Despite challenges, teachers believe AI can broaden students’ horizons.

A recent review of nationwide and global medical data links the shift to an aging population and better detection, with the country's cancer profile changing alongside rising case numbers.

The post Responsible AI Takes Shape appeared first on Partnership on AI .

CSET’s Sam Bresnick shared his expert insight in an article published by The Wall Street Journal. The article examines China’s successful launch and recovery of the Long March-10B rocket booster, a milestone that could significantly reduce launch costs and strengthen China’s ambitions to become a leading space power. The post China Successfully Launches Reusable Rocket in a Win for Space Program appeared first on Center for Security and Emerging Technology .

CSET Executive Director Helen Toner spoke about the global AI competition at the Aspen Ideas Festival. The post Helen Toner Discusses U.S.-China AI Race at Aspen Ideas Festival appeared first on Center for Security and Emerging Technology .

CSET’s Jacob Feldgoise shared his expert insight in an article published by The Wall Street Journal. The article explores the U.S. government’s increasingly active role in supporting domestic semiconductor manufacturing and examines Intel’s efforts to reestablish itself as a leading American chipmaker amid growing competition. The post The White House Made Fixing Intel Its Pet Project. It’s Working. appeared first on Center for Security and Emerging Technology .

On July 9th, President Trump fired the two Democratic commissioners of the U.S. Election Assistance Commission and accepted the resignation of its lone remaining Republican, leaving the agency with zero sitting commissioners four months before the midterms. The White House cited last month’s Supreme Court decision in Trump v. Slaughter, which overturned decades of precedent […] The post After the Firings, Election Officials Need a Plan B for the EAC appeared first on Center for Democracy and Tec

The post The Robotics for Good Youth Challenge 2026 comes to an end as winners are awarded appeared first on AI for Good .

In my first weeks as Executive Director of EFF, I’ve been reminded every day how consequential this moment is in determining what kind of future we will have. We are on the edge. What each one of us steps up to do – with our expertise, energy, and resources – will determine whether our future is one of openness, security, and fundamental rights, or one controlled through fear, surveillance, and centralized power. I am proud to take the torch and help lead our EFF community forward at this pivota

WASHINGTON, D.C. — Last night, federal lawmakers introduced the People-First Chatbot Act, a clear framework to make chatbots safer for everyone and address the harms caused by AI chatbots that were rushed into the public’s hands with little oversight or transparency. EPIC applauds Reps. Valerie Foushee and Greg Casar for sponsoring this important bill.

WASHINGTON, D.C. — Last night, federal lawmakers introduced the People-First Chatbot Act, a clear framework to make chatbots safer for everyone and address the harms caused by AI chatbots that were rushed into the public’s hands with little oversight or transparency. EPIC applauds Reps. Valerie Foushee and Greg Casar for sponsoring this important bill. This legislation was crafted to ensure that the companies operating AI chatbots provide users with clear disclosures, material safeguards, and co

This post is part 2 in a series about automated content moderation. Read the first post here . When whistleblower Frances Haugen leaked a set of documents from Meta in 2020, among the revelations was a jarring statistic: The company’s algorithms designed to detect terrorist content incorrectly deleted nonviolent Arabic-language content 77 percent of the time, while failing to detect hate speech under the company’s own policies in many instances. Meta’s own transparency report released later that

Co-authored with Kelly Brandmeyer, FPF U.S. Policy Intern In a two-day span from June 28 to June 30, the New Jersey legislature introduced and passed A5328, amending New Jersey’s comprehensive privacy law and establishing new data broker and “data collector” registration requirements. The new data broker law has a uniquely broad scope and high financial […]

The Robotics for Good Youth Challenge Grand Finale will reach its conclusion on the Youth Stage from 14:10 to 14:45. After four days of competition, the challenge will come to an end, closing a season that opened the Youth Zone's programme on 7 July. The post Top 3 things happening on Day 4 of the AI for Good Global Summit 2026 appeared first on AI for Good .

In a rare combined cybersecurity/squid post, a twenty-nine-year-old squid proxy bug can leak HTTP requests. As usual, you can also use this squid post to talk about the security stories in the news that I haven’t covered. Blog moderation policy.

HydroShear, a new physics-based simulator, teaches robots how to use their sense of touch to perform complex manipulation tasks, in a way that transfers seamlessly to the real world.

Download the final report here. Artificial Intelligence is showing enormous promise for improving our daily life. Countless applications in many sectors of the economy are already being developed and ...

Last weekend I was in Erfurt, the place where the authoritarian-populist AfD party held its annual federal convention. On Saturday I got up at the crack of dawn to help block the AfD delegates from reaching the assembly hall. Was I supposed to do that? As managing director of a legal-scholarly discourse platform should I not have remained neutral? The demand to forbid oneself from discriminating between non-banned parties loyal to the constitution and non-banned parties hostile to it is question

As I’ve written before, these past couple years I’ve often felt like the last remaining person in either quantum computing or AI who lacked a stake in some startup company whose valuation is right now shooting into interstellar space. My academic colleagues, including the ones who seemed the most singleminded about quantum oracle separations and […]

What Parents of Young Kids Want: Insights from the 2026 National Parent Survey For over 25 years, New America’s insights and policy solutions have helped usher in sweeping, long-term change. Discover ...

This guest post, by Katherine Klosek of ARL and Stephen Wolfson of the University of Pennsylvania, is the latest in […]

Key insights from the latest AI Safety Index | Edition #305

The Summer of Ludd just wrapped in NYC; we talk to a participant and documentarian. Plus, is the Trump admin secretly worried about an AI bubble? The biggest-ever data center gets shut down, and more.

This is part 2 of the weekly, broadly covering speculation, rhetoric and policy, along with alignment research.

In the near future, AI -powered surveillance systems will be able to track everything we do in public, and much of what we do in private. And if we do something wrong—shoplift, litter, jaywalk, you name it—the system will notice, retain it, tie it to your official government record, communicate that fact to you, and provide real-time alerts to any relevant authorities… and maybe also to the general public. Think of these systems as automated speed cameras, but on steroids. Only they’ll enforce n

Today's links "Rights for robots" and the AI slavery fantasy: When we were robots in Egypt… Hey look at this: Delights to delectate. Object permanence: Awkward questions for the entertainment industry; ISP v record label death penalty; Coming out on Splash Mountain; Negative Swiss bond yields; Are We Having Fun Yet? Upcoming appearances: London, Edinburgh, Sydney, Melbourne, Brighton, London, South Bend. Recent appearances: Where I've been. Latest books: You keep readin' em, I'll keep writin' 'e

Microsoft’s latest Secure Future Initiative report outlines progress on secure foundations, AI-powered defense, and future-ready cybersecurity.

Data centers are so hot right now—and not just because of the waste heat. From the halls of Congress to the digital world of podcasts, it seems like all people want to talk about is artificial ...

This paper was accepted at the AI4TCI (Workshop on AI for Secure and Trustworthy Critical Infrastructure Systems) Workshop at the International Conference on Availability, Reliability and Security (ARES) 2026. 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 pri

Commission preliminarily finds the addictive design of Instagram and Facebook in breach of the Digital Services Act Anonymous (not verified) Fri, 07/10/2026 - 10:57 The European Commission has preliminarily found Meta in breach of the Digital Services Act for the addictive design of Instagram and Facebook. The investigation focuses on features such as infinite scroll, autoplay, push notifications, and the platforms' highly personalised recommender systems. The Commission's investigation indicate

The Department of Labor, as part of its continuing effort to reduce paperwork and respondent burden, conducts a pre-clearance consultation program to provide the general public and Federal agencies with an opportunity to comment on proposed and/or continuing collections of information in accordance with the Paperwork Reduction Act of 1995. This program helps to ensure that requested data can be provided in the desired format, reporting burden (time and financial resources) is minimized, collecti

What GAO Found The Department of Veterans Affairs (VA) is working to modernize its IT systems to increase efficiency and accuracy of claims processing for disability compensation. However, past GAO reports have found that VA faces long-standing challenges in managing its disability compensation program and implementing innovative technologies. For example, GAO found gaps in VA’s oversight of the quality of exams provided by contracted medical providers and in the management of training for claim

Nearly 40 global leaders gathered at UNESCO this week for the Transforming Education Summit +4, reaffirming their commitment to SDG 4 as they confront a deepening global education financing crisis and ...

EAP's exports now face higher trade restrictions, elevated policy uncertainty, and slowing global growth. And its workers must cope with the growing use of robots, AI, and digital platforms. Download ...

Industrial sound design requires audio generation systems that not only produce realistic audio, but also preserve the perceptual identity of a reference, support controllable variation, and remain efficient for practical workflows. Existing evaluations are usually tied to text-to-audio (TTA), unconditional, or task-specific settings, limiting assessment for reference-guided sound effects (SFX) variation. To address this gap, we present a production-oriented evaluation framework for structured c

Cybersecurity systems must adapt rapidly to emerging threats. However, labeled data for new threat categories is unavailable when those threats first appear. Generalized zero-shot learning offers a natural solution by enabling recognition of unseen classes through auxiliary semantic knowledge rather than labeled examples. Large language models are particularly promising in this setting because they can convert unstructured CTI reports into semantic prototypes for emerging threats. However, apply

Audio deepfake detection models determine whether speech is genuine or artificially generated, but high overall accuracy can mask substantial performance disparities across demographic groups. In this work, we investigate gender bias in audio deepfake detection using the ASVspoof5 dataset. We use ASVspoof5 under a controlled custom split designed to isolate gender-composition effects. We train attack-specific models on nine training sets with different gender compositions, ranging from female-on

The proliferation of agentic AI systems across enterprise and public-sector contexts has outpaced the capacity of general-purpose AI risk frameworks to classify and govern them. In this paper, we introduce the TrustX Agent Risk Classification Framework, a structured, repeatable instrument that can be applied to seven types of agentic AI systems and is grounded in foundational pre-existing AI governance frameworks. At the core of the framework is a twelve-dimension scoring rubric that robustly qu

Explaining machine-learning models is increasingly important for decision-making and consumer trust, yet it is widely believed to come at a cost: existing Explainable AI (XAI) methods suffer from a persistent accuracy-explainability trade-off. We argue that this trade-off is not fundamental, but an artifact of treating explanation and prediction as separate objectives; when properly coupled, they become complementary, so that equipping a model to explain itself improves, rather than degrades, it

Reinforcement learning (RL) is increasingly used to align multimodal large language models (MLLMs), but higher rewards do not always imply better task performance. This risk is amplified when visual evidence is evaluated by text-only or weakly grounded rewards. We study reward hacking in MLLM RL across safety VQA, chart VQA, and stress-test settings, varying reward design, data ambiguity, model scale (2B-32B), and RL algorithm (GRPO, RLOO, DAPO). We introduce Newly Rewarded Failure Rate (NRFR),

Text-guided medical image segmentation leverages clinical semantics to improve lesion delineation, yet many existing models bind cross-modal fusion, supervision, and decoder design into a task-specific architecture. Such tight coupling makes it difficult to reuse language guidance modules across heterogeneous vision and text backbones, and often requires redesigning the network when the encoder pair changes. This paper presents BTHA, a backbone-transferable hierarchical adapter framework for tex

Long-term animal re-identification (ReID) must remain robust to gradual morphological evolution and seasonal appearance shifts. Although recent vision-language models provide strong pretrained visual representations, adapting them to longitudinal ecological settings remains challenging, particularly under identity and temporal distribution shifts. We present a parameter-efficient CLIP adaptation framework for animal ReID and introduce a continuous metadata-conditioning mechanism that incorporate

Decentralized Autonomous Organizations (DAOs) use token-weighted voting to allocate resources, set protocol rules, and legitimate collective decisions. Yet, support in DAO voting is strikingly concentrated. What happens inside the ballot that produces this concentration? We study DAOs' governance at the proposal-choice level, linking each choice's voting-power share to three observable features: whether it expresses an approval-oriented stance, where it appears in the choice list, and whether it

Ambivalence and hesitancy are subtle behavioral states that are expressed through a combination of verbal content, facial behavior, visual context, and acoustic cues. Effective recognition therefore requires not only extracting informative unimodal representations, but also modeling how temporally aligned behavioral evidence interacts across modalities. In this paper, we propose a synchronized visual-facial cross-refinement framework (SVF-CR) with pairwise multimodal evidence fusion for ambivale

Identifying and assigning keywords at scale is a technical, practical, and ethical challenge for crowdsourced collections. This article reports the findings of the "Extracting Keywords from Crowdsourced Collections" project, which used the Their Finest Hour Online Archive, a crowdsourced Second World War digital collection hosted by the University of Oxford, as a case study. The project evaluated three Natural Language Processing approaches to automate keyword extraction: Named Entity Recognitio

Large language models (LLMs) are increasingly used to summarize and evaluate policy-relevant information, but it remains unclear whether their judgments are implicitly shaped by geopolitical cues. I study this question with an endorsement experiment in which four LLMs evaluate the same international economic and security policies after each policy is randomly described as supported by the United States, the European Union, China, or Russia. In the numeric-only condition, GPT-5, Claude Sonnet, an

Continual learning commonly relies on post-hoc mechanisms such as replay, elastic regularization, or distillation. This work argues that forgetting should instead be modeled directly as interference between tasks. In the frozen-feature regime, forgetting from learning a new task is exactly the interference energy induced on the old task. In deep networks, the same quantity is recovered through path-averaged curvature with minimal additional forward passes. When task supports are disjoint, forget

Recognition tells an agent what is in an image; personal memory affects what is worth looking up next. In a camera-first setting the user can send only an image, so the agent must form the lookups. We study whether personal visual memory improves agent-side tool choice and tool arguments, and thereby more user-aligned multi-tool lookups. The design uses a three-layer personal visual memory (profile, short-term focus, observations) that is loaded on each turn to condition an LLM tool-calling loop

Large language models (LLMs) are increasingly deployed in online medical consultation, yet existing benchmarks remain poorly aligned with real clinical practice. Many rely on synthetic conversations or patient simulators, omit patient-uploaded medical images, or evaluate open-ended clinical responses using multiple-choice or lexical-overlap metrics that poorly reflect clinical quality. We introduce \textbf{MedRealMM}, a large-scale benchmark for multimodal online medical consultation built from

Representation alignment (REPA) has been investigated to accelerate diffusion training, but we observe that regularizing intermediate representations in diffusion Transformers (DiT) may implicitly entangle latents and limit generative capacity. To address this issue, we propose ReGen, a hierarchical multi-prompt representation generation framework that jointly estimates multiple vector fields for both representations and data within a single diffusion model. We further introduce generalized flow

Video anomaly detection (VAD) is critical for automated surveillance but remains fragile under challenging conditions such as illumination variations, fast motion, and complex backgrounds when relying solely on visible light videos. To address these limitations, we propose EVAD, an event enhanced VAD framework that jointly exploits conventional video and event streams captured by bio inspired event cameras. Event sensors asynchronously capture brightness changes with high temporal resolution, of

Medical imaging models are often deployed without the demographic, acquisition, and quality metadata needed for subgroup auditing. Once those metadata disappear, clinically critical failure modes can be masked by strong aggregate performance, and many robust-learning methods lose the group structure they rely on. We present CAPRA, a calibrated proxy-axis framework for hidden subgroup analysis under missing metadata. CAPRA predicts image-derived semantic axes, calibrates axis posteriors on a smal

Driven by next-token prediction, NLP shifted from task-specific models into powerful generalist foundation models. What, then, is the equivalent catalyst needed to achieve a general-purpose model in computer vision? In this paper, we contend that large-scale text-to-video generation serves as a strong pre-training paradigm for computer vision, providing the necessary spatiotemporal priors, vision-language alignment, and scalability required for general visual intelligence. We introduce GenCeptio

Developers' choices about what data a system collects, how it is used and shared, and what defaults govern user choices directly shape users' privacy experiences. Yet, developers often make problematic privacy-related design decisions without realizing the potential consequences. We introduce Privacy Detective, a narrative investigation game that leverages real-world legal documents to train developers' privacy awareness. In the game, players search for privacy violation evidence derived from le

Understanding music requires understanding localized relationships across data modalities, e.g., how time in performance audio maps onto position in a score image. Yet supervision for such local correspondences is difficult to obtain-in practice, we often only have access to coarser global supervision like paired segments of audio and images. To address this gap, we propose FuSiLi (Fused Sinkhorn-Localized Similarity), a similarity score for multimodal contrastive learning operating directly on

Background: Osteogenesis imperfecta (OI) is a rare genetic disorder characterized by bone fragility and recurrent fractures. Emerging biologics demonstrate promise by targeting bone-remodeling pathways, yet evidence for their efficacy and safety remains fragmented and heterogeneous, and no prior systematic review in OI has incorporated artificial intelligence (AI) to synthesize it. Objective: This study aims to systematically evaluate the efficacy and safety of novel biologics in patients with O

Agentic retrieval-augmented generation (RAG) extends static RAG by allowing language models to iteratively reason, generate search queries, retrieve evidence, and predict answers. However, it remains challenging for models to decide when to retrieve, whether to use lexical matching or semantic similarity, and how to control context granularity to prevent irrelevant tokens from interfering with agent reasoning. In this paper, we introduce GRASP, a reinforcement learning (RL) framework for trainin

Reinforcement learning (RL) has become a dominant paradigm for enhancing LLMs' reasoning capabilities. However, RL algorithms with PPO-Clip are inherently limited by exploration collapse. Subsequent works remain primarily heuristic and fail to identify the essential cause of PPO-Clip's failure. This work reveals the fundamental flaw of PPO-Clip: it implicitly measures policy discrepancy using Euclidean metric, which is theoretically inconsistent with the intrinsic geometry on the policy Riemanni

AI coding assistants are now widely used in professional development, yet they offer only limited ways for developers to control how they behave. In this paper, we investigate what kinds of configurations experienced developers want in coding assistants, how they prioritize different types of configuration needs, and which interface mechanisms they prefer. We first synthesize product documentation and prior research on trust and personalization to compile a list of 33 configuration options, grou

Social Media + Society, Volume 12, Issue 3, July-September 2026. This article examines the role the performance-tracking app Strava plays in facilitating professional careers and engagement in a growing endurance sports industry. Through qualitative interviews with 25 elite endurance athletes and their agents, I find ...

Fully autonomous weapon systems need to comply with International Humanitarian Law and underlying ethical principles. This requires the ability to recognize not only objects or persons to be targeted but also protected persons or objects. Such sophisticated object classification abilities, if achievable at all, would have to utilize machine learning techniques. These come with well-known limitations to predictability, reliability and explainability. This article argues such limitations could be

This study introduces a hybrid neuro-symbolic framework that achieves deterministic detection of statutory inconsistency in complex law. We use the U.S. Internal Revenue Code (IRC) for our investigation because its complexity makes it a fertile domain for identifying conflicts. Our research advances a solution for detecting inconsistent provisions by combining Large Language Models (LLMs) with a symbolic reasoner. To evaluate this approach, we conducted experiments using GPT-4o, GPT-5, and Prolo

IntroductionLarge language models (LLMs) can generate persuasive narratives that shift agent behavior in multi-agent systems, but deploying raw, unstructured text as an influence mechanism offers no formal guarantees on effectiveness, interpretability, or controllability.MethodsWe introduce the LLM Influence Compiler (LIC), a solver–critic pipeline that compiles natural-language cooperation directives into structured, parameterized influence policies defined over a five-field schema: network tar

We present RAG-GNN, an end-to-end trainable framework that augments a graph neural network (GNN) encoder for protein interaction networks with a jointly optimized dense retrieval module over a TF-IDF-indexed document corpus, a gated fusion mechanism, and contrastive alignment between node and document representations. The study is positioned as a controlled methodological investigation of whether retrieval augmentation provides measurable benefit beyond a matched GNN-only ablation, rather than a

Background and objectiveArtificial intelligence has improved adenoma detection during colonoscopy, but most models remain black boxes, limiting clinical trust, especially for sessile serrated lesions (SSLs). Concept Bottleneck Models (CBMs) route predictions through human-understandable concepts. We propose an end-to-end CBM for colonoscopic polyp classification with concept-level explanations aligned with clinical standards.MethodsWe defined 62 concepts from NBI International Colorectal Endosco

Large Language Models (LLMs) have demonstrated remarkable proficiency in general-purpose tasks, yet their capacity for fine-grained reasoning in knowledge-intensive domains (KIDs) remains largely unexplored. This study addresses this gap by investigating LLM performance in the specialized field of food chemistry. We introduce a novel benchmark comprising two core tasks: Molecular-to-Food Prediction (MFP) and Food-to-Molecular Prediction (FMP). To support this benchmark, we curated and standardiz

Pricing food products to balance profitability with consumer welfare is a central challenge for retailers. Dynamic pricing is widely used to maximize revenue, yet most pricing models optimize business objectives while overlooking consumer fairness. This paper studies the risk of consumer exploitation under dynamic food pricing in Canada and proposes a methodology that embeds fairness constraints directly into retail sales forecasting. We model total retail trade sales with a log--log Autoregress

Can a small quantum computer forecast a changing signal better than an ordinary classical method? Many studies say yes, but the classical methods they compare against are often left in a basic, untuned state while the quantum model is carefully optimised. We ask what happens when the classical competitor is given exactly the same care: the same size and the same amount of tuning effort. We study two popular reasons a quantum reservoir is thought to help, using exact simulations of small quantum

Machine learning (ML)-based intrusion detection systems (IDSs) are increasingly used to monitor encrypted industrial communication. However, their behavior under realistic private 5G operating conditions remains insufficiently understood. This paper investigates the impact of benign connectivity variations on ML-based IDSs for encrypted Open Platform Communications Unified Architecture (OPC UA) traffic in industrial private 5G networks. Experimental results show that legitimate connectivity even

We study the integration of variational quantum circuits (VQCs) into diffusion models through a squeeze-and-excitation (SE) channel-modulation scaffold that isolates the quantum contribution. Using a role-matched classical control and multi-seed significance testing across DDPM and latent diffusion on MNIST and CIFAR-10, with a score-based NCSN study on MNIST, we find that quantum cores achieve comparable mean FID to the classical control across DDPM and latent diffusion, while paired sampling-s