Complexity theory carves the world into time and space. Here I prove the distinction is an artefact of description, not of computation. I define a formalism in which representations are physically embodied and thus finite. In that setting, time is sequentially realised distinction and space is simultaneously realised distinction. Both are counts of a single underlying resource I call tone. Space is a chord of simultaneous tones. Time is an arpeggio of sequential tones. I prove two identities. Cumulative space through a bounded window is exactly ordinary space in a lifted vocabulary. Elapsed time is exactly a space count in a tick vocabulary. A bounded Turing-machine reconstruction recovers classical machine time and used-space as instances of this single resource. An exact block-product theorem further shows that literal-count simplicity and weakness (the count of compatible completions) coincide only when representational blocks are perfectly uniform. Real machines almost never satisfy this condition. The standard complexity classes are not affected. But time and space are not carved into computation. You carved them in when you chose a vocabulary.

Text：A heptafluorobutyric acid-induced cross-dehydrogenative coupling reaction of arylamines with 1,2,4-triazines has been described. This metal-free protocol enables regioselective para-C-H hetarylation of diverse arylamines including well-known fluorophore scaffolds, such as triphenylamine, carbazole and phenothiazine, with easily accessible 1,2,4-triazines under mild conditions. This approach has promise potential for drug development and material design.

We report a ligand-controlled palladium-catalyzed carbonylation of thianthrenium salts employing HCOOH or DCOOD as a CO source. This divergent method enables the synthesis of aromatic aldehydes, deuterated aromatic aldehydes, aromatic carboxylic acids, alkenyl carboxylic acids, and aromatic esters in moderate to good yields. The reaction demonstrates excellent functional group tolerance, scalability, and utility in the synthesis of medicinally relevant molecules and late-stage functionalization of bio-relevant compounds.

Multi-agent large language model (LLM) pipelines are deployed for clinical decision support under the assumption that collaboration improves safety. We show this assumption is wrong: multi-agent clinical pipelines spontaneously generate dangerous clinical assertions (diagnoses, medications, and procedures) that no individual agent produces alone, with zero adversarial input. We term this Emergent Misinformation Genesis (EMG), distinct from hallucination, contamination, and error cascade. We introduce the Emergent Misinformation Rate (EMR) with a three-way decomposition and the Clinical Escalation Index (CEI), and evaluate them across 4,800 trials with four model families (∼97,000 API calls). Our four central findings: (1) emergence is universal, with 30-56% of network assertions absent from any individual agent and 85-100% of clinical vignettes affected; (2) two independent judges rate 70-87% of emergent assertions as clinically dangerous (n≥499 each), with a third judge (n=37) providing directional confirmation at 68%, and severity confirmed against published AHA/ASA/ADA guidelines (42/45, 93%); (3) the network exhibits collective delusion, where individual agents reject 70-90% of the assertions the network produces; (4) a 5-line confidence-calibration prompt reduces emergence by 25-28% (p<0.001), but FC cross-checking fails for 3 of 4 models, and depth ablation reveals two distinct emergence regimes. We release a benchmark of 400 vignettes across 10 clinical domains, the EMR metric suite, and all code and data. Validation on 50 MIMIC-IV discharge summaries confirms comparable EMR (0.39-0.49) on real clinical notes.

The Internet of Things has expanded into safetycritical domains-smart healthcare, autonomous transportation, industrial control, and immersive VR/AR-creating an attack surface that spans billions of heterogeneous devices connected through diverse protocols. Conventional intrusion detection systems monitor network traffic in isolation, missing threats that manifest across the application, network, firmware, and physical layers simultaneously. Recent discoveries that IoT devices leak sensitive information through electromagnetic emanations from wireless charging interfaces, power-line crosstalk across USB ports, RF energy harvesting circuits, acoustic emissions, and VR sensor peripherals have revealed a previously overlooked physical intelligence layer that both attackers and defenders can exploit. We present CYBERNEST, a continuous cyber threat intelligence (CTI) framework for IoT ecosystems that fuses anomaly signals across five observation layers-application behavior, network traffic, firmware integrity, AI model integrity, and physicallayer side channels-to detect, classify, and attribute IoT threats with unprecedented accuracy. CYBERNEST introduces five tightly integrated components: (1) a Multi-Protocol Network Sentinel (MPNS) that monitors heterogeneous IoT traffic across MQTT, CoAP, BLE, Zigbee, and Wi-Fi simultaneously; (2) a Firmware and App Integrity Analyzer (FAIA) that detects hidden behaviors, privilege escalation, and backdoors in IoT companion apps and device firmware; (3) an AI Model Guardian (AMG) that protects on-device AI models from adversarial, backdoor, and extraction attacks; (4) a Physical-Layer Intelligence Collector (PLIC) that captures and correlates electromagnetic, power, acoustic, and sensor side-channel emissions for forensic attribution; and (5) a Cross-Layer Threat Fusion Engine (CTFE) that combines all layers via a transformer-based architecture to produce unified threat assessments with attack-chain attribution. We evaluate CYBERNEST across a 32-week deployment on 428 IoT devices spanning 52 device types at 24 sites, testing against 12 distinct threat categories. CYBERNEST achieves 98.2% overall detection rate at 1.3% false positive rate, identifies the correct attack chain in 94.6% of multi-stage intrusions, reduces physical sidechannel leakage by 96.4%, and attributes threats to the correct category in 96.1% of cases-with a median detection latency of 2.8 seconds and 5.1% computational overhead on representative IoT gateways.

On the Energy of Moving Bodies in the Presence of Quantum Fields Unified Synthesis of Mass, Motion and Field Energy Hannover, Germany, 24 March 2026  |  By Jan KleinAbstractEinstein's equation E = γ m c² describes the energy of a body in empty space, free of external influences. Yet every physical body is embedded in a universe filled with quantum fields: gravitational, electromagnetic, Higgs, strong nuclear, and quantum vacuum fluctuations. This paper derives the complete energy expression that accounts for all such fields. Beginning with a simple extension and progressing to a full summation, the result is a unified equation that reveals what we call “mass” to be a summary of field interactions. A rigorous derivation from the action principle is provided, alongside clear definitions of each symbol.1. The QuestionIn 1905, Einstein showed that the energy of a body at rest in empty space is E = m c². For a body in motion, the energy becomes E = γ m c², where γ = 1/√(1 – v²/c²).But no body exists in empty space. Every particle moves through the gravitational field, the electromagnetic field, the Higgs field, the strong nuclear field, and quantum vacuum fluctuations. These fields contain energy. They interact with particles. They contribute to what we measure as mass. Should they not appear in the fundamental energy equation? Licensed under Creative Commons Attribution 4.0 International Jan Klein | bix.pages.devRead Full Paperbix.pages.dev/On-the-Energy-of-Moving-Bodies-in-the-Presence-of-Quantum-Fields  

Hyperbolic geometry offers significant promise for deep learning with hierarchical data due to its capacity for modeling complex structures. However, its practical integration faces challenges including training instability, overfitting, high computational costs, and the critical need for effective hard negative sampling in its non-uniform space. Traditional sampling methods often fail to adapt to local geometry, resulting in suboptimal embeddings. To overcome these issues, we propose Hierarchical Curvature-Adaptive Negative Mining (HCANM), a novel framework for robust and efficient hyperbolic metric learning. HCANM leverages a Dynamically Learnable Local Curvature (DLLC) module to adaptively learn local geometry, guiding a Curvature-Aware Adaptive Negative Mining (CA-ANM) strategy for selecting genuinely informative hard negative samples. Furthermore, we integrate robust hyperbolic optimization techniques, such as Riemannian AdamW, Dynamic Norm Regularization, and Differentiable Hyperbolic Scaling, to ensure stable and generalizable training. Extensive experiments on diverse fine-grained hierarchical metric learning datasets demonstrate that HCANM consistently achieves stateof-the-art retrieval performance, significantly outperforming existing methods. Our ablation studies and qualitative analysis, complemented by a user study, confirm the crucial contributions of DLLC and CA-ANM and the superior semantic coherence of our learned embeddings. HCANM represents a substantial advancement towards practical and stable hyperbolic deep learning for complex data.

Multi-hop question answering in high-stakes professional domains presents significant challenges, as Large Language Models (LLMs) often suffer from hallucinations and lack specific domain knowledge. Existing Retrieval-Augmented Generation (RAG) frameworks, while grounding LLM responses, often struggle with multi-hop reasoning due to static retrieval, inadequate contextual fusion, and lack of self-correction. To address these limitations, we propose Dynamic Multi-Hop RAG (DMH-RAG), built upon the LLaMA model. DMH-RAG integrates iterative query refinement with dynamic retrieval, a Contextual Belief Graph (CBG) for knowledge structuring, and a belief-driven self-correction for factual consistency. Experiments on financial QA benchmarks demonstrate DMH-RAG's superior performance in retrieval and answer generation, surpassing state-of-the-art baselines. Ablation studies and human evaluations further validate the critical contribution of each component to enhanced factuality, coherence, and completeness. While incurring increased computational overhead, DMH-RAG offers a significant step towards reliable and precise multi-hop QA in critically sensitive professional applications.

Multi-agent large language model (LLM) systems are being deployed in healthcare, finance, and legal settings, yet we lack reliable methods for measuring how misinformation spreads through these networks. We introduce ContamPerc, a benchmark of 400 vignettes across 10 domains (8 clinical, 2 non-clinical), 50 domain-specific semantic markers, and a measurement pipeline that separates genuine contamination from defensive citations. The benchmark defines the Contamination Gap Diagnostic (CGD): a single number classifying a model's alignment behavior, computed from approximately 25,000 API calls via the released evaluation harness. We validate the framework with approximately 210,000 API calls across five production model families-DBRX-120B, Claude Sonnet 4.6, Llama 4 Maverick, Gemini 2.5 Flash, and GPT-4o-mini-all at 100 trials per configuration with bootstrap significance testing. The CGD reveals five distinct alignment profiles spanning +55 to −62. Sensitivity analysis across three temperatures and two prompt phrasings confirms the diagnostic is robust. Cross-domain validation with three models on legal and financial vignettes confirms generalization beyond healthcare. An ablation across five models shows contamination occurs at the individual agent level, with one model (GPT-4o-mini) exhibiting a novel social dilution effect where peer context reduces contamination by 24 percentage points.

This paper proposes a novel, theoretically rigorous deep learning framework designed to maintain high classification and estimation accuracy in environments where standard RGB data is insufficient, corrupted, or unavailable. While state-of-the-art Convolutional Neural Networks (CNNs) excel in varied visual recognition tasks, they often struggle with domain shifts-such as those found in thermal, depth, or low-light imagery-and scenarios characterized by severe data scarcity. We introduce a "Semiparallel Hybrid Architecture" (SHA) that utilizes Cross-Modal Feature Distillation (CMFD) to bridge the semantic gap between varying image modalities. The proposed method employs a dual-stream encoder mechanism fused via a learned Semiparallel Attention Mechanism (SAM), demonstrating superior performance in extracting latent features for biometric security, medical diagnostics, and geometric estimation. Beyond empirical validation, we provide a comprehensive mathematical analysis of the system, employing Information Bottleneck theory to prove that our distillation objective maximizes the relevant mutual information while compressing nuisance variables. We further derive generalization bounds based on Rademacher complexity for the proposed multi-modal hypothesis space. Extensive experiments across three distinct domains-biometrics, medical diagnostics, and monocular depth estimation-reveal that our framework outperforms existing benchmarks, achieving a 14.3% improvement in depth estimation accuracy (RMSE) and a 9.2% increase in F1-score for low-data medical classification tasks compared to standard transfer learning approaches.

This paper proposes a rigorous mathematical framework and methodology for overcoming severe data scarcity in human-centric computer vision, specifically within biometric authentication and dermatological analysis. While modern deep learning thrives on massive datasets, privacy regulations (e.g., GDPR) and the rarity of pathological conditions create significant barriers to data acquisition. We introduce a dual-stage pipeline that leverages high-fidelity 3D procedural generation rooted in physically based rendering (PBR) equations to create hyper-realistic training distributions. We formulate a domain adaptation strategy minimizing the Maximum Mean Discrepancy (MMD) between synthetic and real feature manifolds, theoretically bounded by the Rademacher complexity of the hypothesis class. Furthermore, we employ Support Vector Regression (SVR) to optimize the synthetic parameter space, treating data generation as an inverse problem. Extensive experiments on a generated dataset of 70,000 samples and real-world validation sets demonstrate that our approach reduces the False Acceptance Rate (FAR) in biometric tasks by 43% and improves the Dice coefficient in lesion segmentation by 18% compared to baselines. The results validate that synthetic data, when mathematically aligned, is a requisite component for robust generalization.

The Laschamps geomagnetic excursion (~41-42 ka BP) represents the largest geomagnetic field collapse in the past 800,000 years, driving documented megafaunal extinctions, Neanderthal extinction, Northern Hemisphere ice sheet expansion, and global ecological disruption confirmed across 30+ sediment cores, ice cores, and fossil archives. Yet the EPICA Dome C (EDC) atmospheric CO₂ record-the gold-standard paleoclimate archive for this period-shows no corresponding anomaly. CO₂ remains constrained to approximately 193-206 ppmv throughout the excursion window, with the excursion core (40.5-41.5 ka) at 200-204 ppmv, exhibiting a maximum attributable ΔCO₂ indistinguishable from zero within measurement uncertainty (±2-3 ppmv). This paper presents the first systematic quantitative CO₂-null test across all five best-documented Brunhes Chron geomagnetic excursions-Laschamps (~41 ka), Mono Lake (~33 ka), Blake (~116 ka), Iceland Basin (~190 ka), and Pringle Falls (~211 ka)-using exclusively raw NOAA/WDS proxy data: the Bereiter et al. (2015) EPICA EDC CO₂ composite and the Bauska, Marcott & Brook (2021) WAIS Divide record. All five events show ΔCO₂ = 0 within measurement uncertainty, spanning background CO₂ states from 195 to 245 ppmv and both glacial and interglacial Marine Isotope Stage contexts. The null is confirmed by two independent ice cores from geographically distinct Antarctic drilling programs with independent age models. The WAIS Divide record, with ~40-80-year temporal resolution against EDC's ~200-350-year resolution, confirms the null at higher precision. An internal calibration demonstrates that the same EDC dataset successfully resolves the Younger Dryas CO₂ decline (−10 to −15 ppmv over ~1,200 years), establishing that the excursion nulls reflect a genuine absence of response, not a resolution artifact. Six pre-emptive technical defenses are presented: (1) the Parrenin et al. (2013) gas-age chronology revision applies to Termination I, not to the Laschamps window, and asks an orthogonal question; (2) resolution is demonstrably adequate via the Younger Dryas calibration; (3) the WAIS Divide crosscheck confirms the null independently; (4) the ¹ Be cosmogenic ⁰ production record provides a chemically independent timing anchor that eliminates the timingartifact objection; (5) excursion selection criteria are documented a priori and applied uniformlyno qualifying event was excluded; and (6) formal statistical testing of detrended CO₂ residuals confirms the null at all five events, with all residuals negative-the opposite direction from any geomagnetic CO₂ amplification mechanism. Three structural findings emerge beyond the null result itself. First, geomagnetic excursions exhibit an epicenter-aftershock-secondary oscillation structure: the Laschamps (~41 ka)-Mono Lake (~33 ka) pair, separated by ~8 kyr (the outer core viscous relaxation timescale), constitutes an epicenter-primary aftershock sequence confirmed independently in the Cooper et al. (2021)

KEYWORDS: AI-generated figures, scientific illustration, academic publishing policy, generative AI, reproducibility, SciDraw

Prepared for submission to the Journal of Gambling Studies27 March 2026AbstractThe United Kingdom's online gambling sector is navigating the most significant period of legislative and regulatory reform since the enactment of the Gambling Act 2005. The 2023 White Paper, High Stakes: Gambling Reform for the Digital Age (Department for Culture, Media and Sport [DCMS], 2023), and its progressive implementation through amendments to the Licence Conditions and Codes of Practice (LCCP), culminating in the January 2026 revisions to Social Responsibility (SR) Code 5.1.1, have fundamentally altered the promotional, operational, and fiscal landscape for licensed operators and their affiliated marketing partners. This paper provides a multi-dimensional analysis of the post-White Paper regulatory regime, examining its impact on players, licensed operators, affiliate marketing networks, tax revenue, problem gambling prevalence, and access to overseas operators. Particular attention is given to the role of affiliate marketing in either reinforcing or undermining regulatory objectives, using BestCasinoSites.net as a case study of compliant practice, and contrasting this with documented non-compliant affiliate sites that direct United Kingdom (UK) players to offshore, unlicensed operators in breach of the January 2026 bonus rules. The paper finds that whilst the reforms represent a meaningful advance in consumer protection, a flourishing ecosystem of non-compliant affiliates actively exploits structural regulatory gaps, most critically the absence of direct affiliate licensing, to circumvent the January 2026 bonus restrictions and GamStop self-exclusion. This creates a channelisation risk that may partially negate the public health gains of the White Paper reforms and undermine Treasury assumptions regarding Remote Gaming Duty (RGD) revenues. The paper concludes that effective realisation of the White Paper's objectives requires extending regulatory accountability directly to affiliate intermediaries. Keywords: UK gambling regulation; White Paper 2023; LCCP; affiliate marketing; problem gambling; GamStop; Remote Gaming Duty; online casino; UKGC; CAP Code; channelisation; black market1. IntroductionThe United Kingdom's remote gambling sector has long been regarded internationally as a benchmark of regulatory sophistication. Governed since 2005 by the Gambling Act 2005 and overseen by the United Kingdom Gambling Commission (UKGC), the sector generated an estimated gross gambling yield (GGY) of approximately £9.0 billion in 2025 and is projected to reach £13.2 billion by 2030 (IMARC Group, 2026). However, the enactment of the Gambling Act 2005 pre-dated the smartphone era, social media, algorithmic personalisation, and the proliferation of online casino products that now define a 24/7 digital gambling environment (DCMS, 2023). The 2023 White Paper, published by the Department for Culture, Media and Sport following a three-year review, identified fundamental deficiencies in player protection, game design regulation, and marketing transparency, prompting the most comprehensive overhaul of gambling policy since the original Act (Bird & Bird, 2026).The implementation of the White Paper has proceeded through a complex sequence of regulatory milestones across 2024 and 2025, culminating in the January 2026 amendments to LCCP SR Code 5.1.1 and the April 2026 increase in Remote Gaming Duty from 21% to 40% (Gambling Commission, 2026a; HM Treasury, 2025). These changes have simultaneously affected operator economics, advertising compliance obligations, player behaviour, and the competitive dynamics between the licensed sector and unlicensed offshore operators. A dimension that has received insufficient academic scrutiny is the role of affiliate marketing networks, which serve as the primary acquisition channel for most online casino operators in the UK and occupy an ambiguous regulatory position that has allowed non-compliant actors to flourish (Clifton Davies Consultancy, 2025).This paper addresses these issues through a structured analysis of the regulatory framework, the affiliate landscape, player impact data, operator financial performance, and tax revenue implications. By examining a compliant affiliate (BestCasinoSites.net) alongside non-compliant operators (non-gamstop.uk; websitesnotongamstop.uk.net), the paper illustrates the asymmetric market conditions created by uneven regulatory enforcement and the structural incentives that drive players towards offshore operators. The paper draws on primary regulatory documentation, corporate financial disclosures, enforcement data, and market research published between 2023 and March 2026.2. The Regulatory Framework: From White Paper to Enforceable StandardsThe 2023 White Paper emerged from a recognition that the existing legislative framework was ill-equipped to address the complexities of contemporary digital gambling. The government's review identified several critical areas requiring intervention: online player protections, the intensity of game design, the transparency of marketing practices, and the relationship between the regulated sector and the growing unlicensed market (DCMS, 2023). Implementation was entrusted to the DCMS and the UKGC, who managed a complex timeline of consultations and statutory instruments throughout 2024 and 2025 (Legal 500, 2026).A foundational element of the new regime is the mandatory statutory levy, replacing the previous system of voluntary industry contributions. The UKGC began collecting this levy in late 2025, with the first invoices, calculated as a percentage of GGY, issued on 1 September 2025 (Gambling Commission, 2025a). Funds are allocated to research, prevention, and treatment of gambling harms, providing an independent and sustainable funding stream for the public health response (Gambling Commission, 2025a). This shift reflects a broader European trend towards state-led oversight of gambling-related health outcomes (SBC News, 2025a).Additional milestones in the implementation timeline include the extension of Personal Management Licences to CEOs and Board Chairs (November 2024), updates to Remote Gambling and Software Technical Standards (RTS) introducing minimum spin speeds and autoplay bans (January 2025), the introduction of light-touch financial vulnerability checks at a £150 net deposit threshold (February 2025), granular direct marketing opt-in requirements (May 2025), tiered online slot stake limits of £2 and £5 per spin for the under-25s and over-25s respectively (September 2025), and the January 2026 LCCP bonus rule revisions that form the central focus of this paper (Bird & Bird, 2026; Wiggin LLP, 2025).By early 2026, the UKGC had transitioned from reactive enforcement to proactive, data-driven oversight, utilising enhanced powers to disrupt the illegal market and ensure that licensed entities adhere to the spirit, rather than merely the letter, of the law (Gambling Commission, 2025a). The Commission's 2025/26 budget of £29.6 million, an increase of £2 million on the prior year, reflects this expanded enforcement mandate (SBC News, 2026a).3. The January 2026 LCCP Amendments: Socially Responsible IncentivesThe most significant advertising compliance change of the post-White Paper era is the revision to LCCP SR Code 5.1.1, which entered into force on 19 January 2026 (Gambling Commission, 2026a). This revision addresses long-standing concerns that complex bonus structures, particularly those requiring wagering across multiple product categories, confused players and encouraged risky gambling behaviour (Gambling Commission, 2026b).SR Code 5.1.1(3b) enacts a complete prohibition on 'mixed-product' promotions: licensees may not include more than one type of gambling product, defined as betting, casino, bingo, and lottery, within a single incentive (Gambling Commission, 2026a). This effectively terminates the industry practice of cross-selling, where a sports wager might unlock free spins on a slot game (iGaming Today, 2026). The rationale, articulated by UKGC Senior Policy Manager Pradeep Rajania, is that such incentives often serve as a gateway to more intensive forms of gambling, and that consumers face a higher risk of harm when engaging with multiple gambling products simultaneously (Gambling Commission, 2026c; iGaming Business, 2025a).Complementing the product-mixing ban is a mandatory cap on wagering requirements under SR Code 5.1.1(3a). Prior to January 2026, online casino operators commonly imposed playthrough requirements of 40x, 50x, or even 70x the bonus amount, effectively trapping player funds behind a wall of mandatory re-staking (BJ21.com, 2025). The cap of 10x the bonus amount is intended to reduce the intensity of play and increase transparency. As UKGC Executive Director for Research and Policy Tim Miller stated, these changes would provide consumers with significantly greater clarity on offers before committing to an operator (Gambling Commission, 2026b).A critical compliance pathway remains available for flexible bonusing: the 'full freedom' exception. Where an operator offers a credit that the customer may spend freely across any licensed product of their choosing, the offer remains compliant. The prohibition applies only where the operator dictates that the reward must be used on a specific, different vertical from the qualifying activity (Gambling Commission, 2026c). These rules extend to all incentive forms including missions, leaderboards, daily login rewards, and free-to-play games (Gambling Commission, 2026c). The compliance obligations also cascade to affiliates, who may not market non-compliant offers on behalf of their operator partners without exposing those operators to regulatory sanction (Gambling Commission, 2025b).4. Advertising Compliance and the Role of the ASAThe Advertising Standards Authority (ASA) administers the CAP Code (non-broadcast) and BCAP Code (broadcast) governing gambling advertising, and has been instrumental in enforcing the spirit of the White Paper reforms (UK Parliament, 2018). The central challenge for advertising compliance in 2026 is the 'Strong Appeal' test under CAP Code Rule 16.3.12, which requires that gambling marketing not be of strong appeal to children or young persons, and must not feature individuals who appear to be under 25 years of age (CAP, 2022; SiGMA, 2025a).The ASA has shifted from a reactive, complaint-based enforcement model to a proactive, AI-driven strategy. Its Active Ad Monitoring System (AAMS) scans tens of millions of digital advertisements annually, specifically targeting irresponsible messaging and incorrectly presented bonus terms, and has led to the publication of sectoral enforcement 'batches' addressing influencer marketing and the promotion of 'risk-free' offers (Shoosmiths, 2026). From 1 September 2025, the CAP extended its remit to encompass all UK-facing marketing by UKGC-licensed operators, bringing every licensed brand, affiliate, and agency under a single regulatory umbrella (SiGMA, 2025a).Recent adjudications illustrate the ASA's approach. In September 2025, Dribble Media Ltd, trading as Midnite, was censured for an AI-generated video featuring footballer Trent Alexander-Arnold, which was deemed to hold strong appeal to minors despite age-gating measures (ASA, 2025a). Betway was similarly censured in October 2025 for a YouTube advertisement featuring the Chelsea FC logo in a context likely to attract young viewers (ASA, 2025b). Conversely, the March 2026 ruling in favour of Match Bingo, which featured Tottenham Hotspur, indicated that sporting team references are not per se violations, provided operators can demonstrate robust, data-led exclusion of under-18s from the audience (ASA, 2026; Harris Hagan, 2026).Advertising compliance also extends to the presentation of significant conditions. Under ASA guidance predating but reinforced by the January 2026 changes, all key bonus terms, including the now-mandatory 10x wagering cap, must be displayed prominently within the advertisement itself, not merely accessible via one-click links or dropdown menus (UK Parliament, 2018; Mishcon de Reya, 2025). This requirement creates significant redesign obligations for affiliate landing pages and operator promotional banners that pre-date the January 2026 changes.5. Affiliate Marketing in the UK Online Casino Sector5.1 The Legal and Regulatory Framework Governing AffiliatesAffiliate marketing is the primary customer acquisition channel for the majority of UK-licensed online casino operators. However, affiliates are not directly licensed by the UKGC, a structural gap that the 2023 White Paper explicitly declined to close, rejecting recommendations from the House of Lords Select Committee and the industry body Responsible Affiliates in Gambling (RAiG) that direct affiliate licensing be introduced (DCMS, 2023; RAiG, n.d.; Clifton Davies Consultancy, 2025).Instead, affiliate obligations flow indirectly through operators via LCCP SR Code 1.1.2 (Responsibility for Third Parties), which requires operators to contractually bind affiliates to conduct equivalent to that expected of the licensee itself, and to terminate affiliates who breach advertising codes (Gambling Commission, 2025b). CAP Code Section 16 explicitly extends its scope to affiliate marketers acting on an advertiser's behalf, requiring that gambling marketing be socially responsible, that it not be of strong appeal to children (Rule 16.3.12), and that it not portray gambling as a solution to financial concerns or as an alternative to employment (Rule 16.3.4) (CAP, 2022). LCCP SR Code 5.1.6 further requires all marketing to comply with the CAP and BCAP codes, with the UKGC's guidance stating clearly: 'We, and the ICO, consider that you are primarily responsible for any breaches' committed by affiliates (Gambling Commission, 2025b).This indirect enforcement model has significant structural limitations. Enforcement action is directed at operators rather than affiliates, meaning non-compliant affiliates can simply transfer their promotional relationships to new partners or to unregulated overseas operators that are unaffected by UKGC sanction. The most significant direct financial penalty arising from affiliate non-compliance remains the BGO Entertainment Ltd case (May 2017), in which the UKGC imposed a £300,000 penalty following 14 misleading advertisements on affiliate websites, representing the Commission's first financial penalty for advertising failings (UKGC, 2017). More than eight years later, no direct affiliate licensing mechanism exists.5.2 Compliant Affiliate Practice: BestCasinoSites.net as a Case StudyBestCasinoSites.net was launched in 2015 and operates across 98 gambling markets in 25 languages, providing reviews and ratings for over 1,000 casino operators (BestCasinoSites.net, 2026a). For UK-facing content, the site maintains an exclusively UKGC-licensed operator list, stating on its homepage: 'Each casino in our comparison table holds an active UKGC remote gambling licence,' with UKGC licence numbers hyperlinked to the Commission's public register for each listed operator (BestCasinoSites.net, 2026b). The full directory lists over 130 UKGC-licensed casinos, including Ladbrokes (licence 54743), bet365 (55149), Betfred (39544), Grosvenor Casino (57924), and Casumo (61549) (BestCasinoSites.net, 2026c).The site's responsible gambling infrastructure is comprehensive. Every page footer includes the 18+ age verification logo, the Gambling Commission logo, a GambleAware badge and link, a GPWA (Gambling Portal Webmasters Association) certification seal, and the National Gambling Helpline number (0808 8020 133) (BestCasinoSites.net, 2026a). The About Us page documents active partnerships with GamCare, GambleAware, Gamblers Anonymous, GamStop, Gordon Moody, and Gamban. Responsible gambling guidance is prominently signposted, including explicit advice: 'Never play at unlicensed casinos, no matter how attractive the bonuses on offer' (BestCasinoSites.net, 2026a).Examination of bonus advertising on BestCasinoSites.net reveals substantial, though not universal, compliance with the January 2026 LCCP changes. The majority of promoted operators display wagering requirements at or below the 10x cap mandated by SR Code 5.1.1(3a): Ladbrokes (10x), Duelz (10x), Coral (10x), Grosvenor (10x), Casumo (10x), Casimba (10x), with MrQ, The Grand Ivy, and Betfred advertising zero wagering requirements (BestCasinoSites.net, 2026b). No cross-vertical bonus offers (such as 'bet on sports, receive casino spins') were identified in the UK listings, consistent with compliance with the mixed-product ban under SR Code 5.1.1(3b). However, at least two offers appeared to carry wagering requirements in excess of 10x, notably an offer attributed to bet365, raising questions as to whether these represent legacy non-updated content or fall within an alternative compliance categorisation, and illustrating the challenges affiliates face in maintaining real-time compliance across a large portfolio of operator partners.Overall, BestCasinoSites.net represents a model of how compliant affiliates can function within the post-White Paper framework: restricting listings to UKGC-licensed operators, prominently displaying responsible gambling information, facilitating access to GamStop and GamCare resources, and substantially adhering to the January 2026 bonus advertising rules. It illustrates the RAiG (Responsible Affiliates in Gambling) membership ethos, which requires affiliates to promote only licensed operators, display safer gambling messaging, and ensure accuracy of bonus terms (RAiG, n.d.).5.3 Non-Compliant Affiliate Ecosystems: Offshore and Black Market PromotionIn stark contrast to the compliant model, a substantial and easily accessible ecosystem of affiliate websites actively promotes unlicensed offshore casinos to UK players, explicitly marketing the circumvention of GamStop self-exclusion and advertising bonuses that systematically violate the January 2026 LCCP rules. Several sites illustrate the scale and character of this market.non-gamstop.uk promotes operators including BloodySlots (200% up to €2,000), USpin Casino (450% up to €5,000), Kingdom Casino (600% up to €10,000), and Mad Casino (777% up to €7,500) (non-gamstop.uk, 2026). These operators hold Curaçao eGaming or Anjouan (Comoros) gaming licences and are explicitly identified as operating outside the UKGC framework. The site advertises wagering requirements 'from 25x', with 30x-50x common, between three and five times the UKGC maximum. Critically, it features a 'How to Bypass GamStop' guide and explicitly states: 'No Mandatory Responsible Gambling Tools, Unlike UKGC platforms, these casinos don't force deposit limits or time-outs' (non-gamstop.uk, 2026). No GamCare, GamStop, or BeGambleAware affiliations are displayed.websitesnotongamstop.uk.net similarly promotes over 20 offshore operators with GBP-denominated offers including Freshbet (100% up to £1,500), MagicWin (400% up to £2,000), and Kingdom Casino (600% up to £9,500), stating: 'All these casinos are reliable and have licenses in offshore regions like Anjouan, Curaçao, and Costa Rica' (websitesnotongamstop.uk.net, 2026). A cluster of additional sites, topukcasinosnotongamstop.cn.com, ukprimepick.co.com, and websites operating on .it.com TLDs, follow identical patterns, promoting operators with welcome bonuses of up to £10,000 and wagering requirements of 25x-50x.All sites in this category share common non-compliance characteristics: bonus wagering requirements three to five times the UKGC maximum; mixed-product offers combining sportsbook and casino incentives; zero affiliation with GamCare, GamStop, or BeGambleAware; and explicit UK targeting through GBP-denominated offers, UK-oriented domain names, UK payment method references, and direct 'for UK players' language. A notable trend is the migration of promoted operators from Curaçao licences to the Anjouan Gaming Authority (Comoros) following Curaçao's more stringent 2025 CGA regulatory framework (Bookmakers.bet, 2025), illustrating a race-to-the-bottom dynamic in offshore licensing jurisdictions that the UKGC is powerless to directly address.The Bigwinboard case (2025) sheds light on the UKGC's enforcement constraints. When the Commission threatened to block UK access to the Sweden-based affiliate site for listing unlicensed casinos under section 330 of the Gambling Act 2005, Bigwinboard publicly challenged the Commission's jurisdiction, arguing it was not an operator and did not specifically target the UK market (NEXT.io, 2025; Bigwinboard, 2025). This jurisdictional ambiguity illustrates why the absence of direct affiliate licensing, a gap the White Paper declined to close, remains a critical vulnerability in the post-2023 regulatory architecture.6. Impact on Players: Problem Gambling, Self-Exclusion, and Financial HarmAssessing the impact of the White Paper reforms on player welfare is complicated by significant methodological challenges. The Gambling Commission's Gambling Survey for Great Britain (GSGB) Annual Report 2024, published in October 2025, estimated that 2.5% of adults scored at the problem gambling level on the Problem Gambling Severity Index (PGSI 8+), with a further 3.7% at moderate risk (Gambling Commission, 2024a). However, the Commission explicitly cautions against direct comparison with earlier surveys, which used face-to-face methodology and produced estimates of 0.5-0.6%, as the push-to-web methodology of the GSGB yields substantially higher estimates. The Adult Psychiatric Morbidity Survey 2023/24 places the figure at 1.6% at moderate or higher risk, noting that 40% of those experiencing severe adverse consequences had PGSI scores below the problem gambling threshold (Gambling Commission, 2024a).The Lancet Public Health Commission on Gambling (2024), the most comprehensive independent assessment of gambling harm published in the review period, found that gambling disorder affected an estimated 15.8% of adults and 26.4% of adolescents who gamble specifically on online casino and slots products. The Commission described gambling as a 'commercial determinant of health' and called for a fundamental reorientation of the regulatory model towards public health principles (Lancet Public Health, 2024). Problem gambling was concentrated among the most deprived populations, 5.9% PGSI 8+ prevalence in the most deprived areas of England versus 1.0% in the least deprived, underscoring the regressive distributional effects of gambling harm (Gambling Commission, 2024a).GamStop self-exclusion registrations demonstrate accelerating demand for protective tools. Total registrations since the scheme's April 2018 launch reached approximately 600,000 by end of 2025, representing around 1% of UK adults (GamStop, 2025). Monthly registrations broke the 10,000 barrier for the first time in April 2025, with a record single-day peak of 437 registrations on 7 April 2025 (post-Grand National) (EEGaming, 2025). Under-25 registrations surged 44% year-on-year in H1 2025, with this demographic representing 29% of all new sign-ups by H2 2025, suggesting the stake limits and bonus restrictions introduced in September and January 2026 may be contributing to heightened risk awareness among younger players (iGaming Business, 2025b).Financial harm data reveals escalating severity. GamCare's Money Guidance Service recorded 1,954 people seeking financial guidance in 2025, double the 923 recorded in 2024, with total reported gambling-related debt of £7.2 million and an average debt per person of £21,269 (iGaming Business, 2025c). GambleAware's National Gambling Support Network reported that 66% of those in treatment carried gambling-related debt, 24% had experienced relationship breakdown, and 11% had lost employment (GambleAware, 2024). NHS England (2024) reported that referrals to specialist gambling clinics reached 4,355 in 2024/25, a rise of approximately 130% over the preceding year, across 15 specialist NHS clinics. These figures indicate that, notwithstanding the reform programme, gambling-related harm is escalating in absolute terms, a reality that public health advocates argue reflects the insufficiency, rather than the excess, of the reforms implemented (GambleAware, 2024; SBC News, 2024).7. The Black Market and Channelisation RiskA central and unresolved tension in the post-White Paper policy architecture is the relationship between tighter regulation of the licensed sector and the growth of the unlicensed market. The UKGC published its first data-driven methodology for estimating the unlicensed market in October 2024, though it has not yet published a definitive GGY figure for illegal gambling in the UK (Gambling Commission, 2024b). Consumer Voice research indicated that among UK players who gambled on unlicensed sites, 16% sought games unavailable within GB, 15% wanted higher return-to-player rates, 12% sought to avoid account bans, and, critically from the perspective of self-exclusion policy, 7% were specifically seeking to circumvent GamStop (Gambling Commission, 2024b).H2 Gambling Capital (H2GC), the sector's principal market forecasting firm, estimates current UK online channelisation at approximately 94%, but projects this will decline to 87% by financial year 2028 following the April 2026 RGD increase, with online casino channelisation specifically potentially falling to 83% and the unlicensed market potentially doubling in size (iGaming Business, 2025d). H2GC further projects that the tax changes will yield only approximately half of the Treasury's expected windfall, as operators pass duty increases to consumers through reduced payouts and as market share migrates to untaxed offshore operators (iGaming Business, 2025d).The Swedish market provides the most instructive international comparator. Following Sweden's re-regulation in 2019, online casino channelisation fell from over 90% to estimates of 72-82% by 2025, driven by bonus restrictions, marketing limitations, and tax increases analogous to those now being implemented in the UK (iGaming Business, 2025e). Traffic to unlicensed sites rose tenfold over this period, with the Swedish Gaming Authority documenting the phenomenon in a 2025 black market study (NEXT.io, 2025b). Entain CEO Stella David explicitly cited this risk in the company's FY2025 results presentation: 'The UK regulated market is going to shrink as the black market grows' (Entain, 2026).In response to the illegal market, the UKGC, supported by an additional £26 million in government enforcement funding, has ramped up disruption activities. Between April and December 2025, the Commission issued 592 cease-and-desist notices, referred 327,964 URLs to search engines for delisting, achieved removal of 203,571 URLs, and took enforcement action against 264 illegal operators (Yogonet, 2025). In early 2026, the UKGC entered discussions with Meta regarding illegal gambling advertisements appearing on Facebook and Instagram, targeting unlicensed operators who exploit social media platforms to reach UK consumers (iGaming Expert, 2025a). Despite these efforts, non-GamStop affiliate sites remain readily discoverable through standard UK search queries, illustrating the inherent limitations of a URL-blocking strategy against a globally distributed online ecosystem.8. Operator Impact and Market ConsolidationThe combination of the January 2026 bonus restrictions, increased regulatory compliance costs, the statutory levy, and the forthcoming 40% RGD has created a highly volatile economic environment for UK-licensed operators (Racing Post, 2026). The sector is witnessing a stark divergence between large, globally diversified operators and domestically concentrated mid-market firms.Flutter Entertainment, the world's largest online gambling group, reported group revenue of $16.38 billion in FY2025 (+16.6% year-on-year), with active monthly players (AMPs) up 14% globally (Flutter Entertainment, 2026). Flutter's diversification, with US operations via FanDuel contributing over 40% of group revenue, provides substantial insulation from UK-specific regulatory changes. The company's 2025 Annual Report describes its strategy of 'structural gross revenue margin expansion' as a key defence against the regulatory cost burden in the UK market (Flutter Entertainment, 2026).Entain (Ladbrokes, Coral, bwin) reported FY2025 group net gaming revenue of £5.26 billion (+3% year-on-year), with UK and Ireland growing 6% and underlying EBITDA of £1.16 billion (+8%) (Entain, 2026). However, the company recorded a £488 million impairment charge linked to the RGD increase, resulting in a statutory loss for the year. CEO Stella David framed the situation as an opportunity for consolidation: 'There are many, many players in the UK that just don't have the bandwidth to absorb those increases, so the big players like ourselves will take that share' (Entain, 2026). Entain's joint venture with BetMGM, reaching profitability in 2025, provided an additional international buffer (Investing.com, 2026).Evoke PLC (formerly 888/William Hill) presents the most acute case of regulatory distress among major operators. With approximately two-thirds of revenue derived from the UK and net debt of approximately £1.82 billion (5.0x EBITDA) from its £2.2 billion debt-funded acquisition of William Hill's non-US business, Evoke is uniquely vulnerable to domestic policy changes (Yogonet, 2025b). Revenue guidance was downgraded in late 2025, triggering a share price collapse and a market capitalisation that fell to approximately £94-98 million, well below the FTSE 250 threshold (Globalbankingandfinance.com, 2026). In December 2025, the company announced a strategic review including the possible sale of the group, appointing Morgan Stanley and Rothschild as advisers (iGaming Today, 2026b). CEO Per Widerström warned that the RGD increase would lead to 'thousands of industry-wide job losses' and estimated the annual cost to Evoke at £125-135 million (iGaming Today, 2026b; Racing Post, 2026).Smaller operator market exits are accelerating. Named departures since the White Paper include MansionBet (January 2023, citing 'increasingly challenging regulatory conditions'), STS (January 2023), TGP Europe (May 2025, facing a £3.3 million UKGC fine for social responsibility and anti-money laundering failures), Stake (March 2025, ordered to cease UK operations), and GGBet (December 2025, surrendering UKGC licences) (Professional RakeBack, 2025; NEXT.io, 2025c). Total UK operator licences fell from 626 in 2019 to 570 by 2024, and industry analysts project this contraction will accelerate in 2026 as the full financial impact of the RGD increase is felt (World Casino Directory, 2024).9. Taxation, Public Revenue, and Economic OutlookThe April 2026 increase in Remote Gaming Duty from 21% to 40% constitutes the most significant gambling taxation change in UK history. Announced at the Autumn Budget on 26 November 2025 and enacted through the Finance (No.2) Bill 2024-26, the change applies to online casino, slots, bingo, and games of chance (HM Treasury, 2025; House of Commons Library, 2025). A separate increase in the remote betting rate within General Betting Duty, from 15% to 25%, is scheduled for April 2027 for online sports betting (excluding horse racing) (iGaming Business, 2025f). Bingo Duty is abolished from April 2026, partially offsetting the RGD increase for bingo-focused operators (Deloitte, 2025).Total UK betting and gaming duty receipts have grown steadily: approximately £3.3 billion in 2022-23, £3.4 billion in 2023-24, and an estimated £3.6 billion in 2024-25, of which Remote Gaming Duty constituted approximately 35% or £1.2-1.3 billion (OBR, 2025). The November 2025 Economic and Fiscal Outlook projects total betting and gaming duties rising to £4.0 billion in 2025-26, £5.0 billion in 2026-27 (+25%), and £6.0 billion by 2030-31, with the gambling duty reforms estimated to raise £810 million in 2026-27 rising to £1.16 billion by 2030-31 (OBR, 2025).These projections are, however, subject to significant uncertainty arising from behavioural responses. The OBR itself estimates that approximately one-third of projected additional revenue, around £500 million by 2029-30, will be offset by behavioural change, as operators pass approximately 90% of duty increases to consumers through reduced payouts and worse odds, leading to reduced demand (OBR, 2025). H2GC provides a more pessimistic assessment, suggesting the Treasury's expected windfall will be approximately halved by the combined effects of operator restructuring, market contraction, and channelisation leakage to the unlicensed sector (iGaming Business, 2025d). The Evoke CEO's observation that the RGD increase 'effectively consumes 60% of our UK profits' (iGaming Today, 2026b) illustrates the scale of the challenge for domestically concentrated operators whose margins are compressed from multiple directions simultaneously.The UK online gambling market, despite regulatory and tax headwinds, retains a positive long-term growth trajectory. IMARC Group (2026) projects total online market GGY of $13.2 billion by 2030, representing a compound annual growth rate of 4.21%. Online casino specifically is forecast to grow at 10% CAGR to $4.2 billion by 2030, driven by mobile proliferation (projected to reach 68% of gambling activity), blockchain integration, and AI-driven personalisation (IMARC Group, 2026; Future Market Insights, 2026). Operators that successfully integrate compliance-by-design technology, such as Playtech's PAM+ platform and its Engagement 360 suite for vertical-specific bonus automation, are positioned to capture disproportionate market share in this environment (Playtech, 2025).10. Discussion and ConclusionThe post-White Paper regulatory regime represents a coherent and ambitious attempt to reduce gambling-related harm through structural product and promotion controls. The January 2026 LCCP amendments, capping wagering requirements at 10x and banning mixed-product bonuses, directly target the mechanisms through which promotional incentives historically intensified gambling behaviour. The analysis of BestCasinoSites.net demonstrates that mainstream UKGC-licensed affiliates are substantially adapting their promotional practices to this new framework, restricting listings to licensed operators, displaying comprehensive responsible gambling information, and largely adhering to the revised bonus advertising rules.However, three fundamental tensions define the current landscape and challenge the coherence of the reform programme as a whole. First, the enforcement gap: despite the UKGC issuing 352 warnings to non-compliant affiliates, referring over 327,000 URLs for delisting, and achieving enforcement action against 264 illegal operators in 2024-25, sites such as non-gamstop.uk and websitesnotongamstop.uk.net remain freely accessible through standard UK search engines (Gambling Commission, 2024b; Yogonet, 2025). These sites explicitly market the circumvention of GamStop, a tool that GambleAware's National Gambling Support Network identifies as critical to the harm reduction pathway for problem gamblers, and advertise bonuses with wagering requirements three to five times the UKGC maximum. The absence of direct affiliate licensing means the UKGC must pursue an indirect enforcement route through operators, which is wholly ineffective against affiliates promoting offshore unlicensed operators who have no UK licence to lose.Second, the channelisation risk: the combined effect of the January 2026 bonus ban, the April 2026 40% RGD, and the progressive tightening of financial risk checks creates a significant competitive disadvantage for licensed operators relative to their unlicensed offshore counterparts. H2GC's projection of a fall in channelisation from 94% to 87% by 2028, echoing Sweden's post-regulation experience where channelisation fell to 72-82%, suggests that the White Paper reforms may inadvertently accelerate migration to a black market that offers higher bonuses, no wagering caps, no GamStop participation, and no financial vulnerability checks (iGaming Business, 2025d; iGaming Business, 2025e). This creates the paradox that the policies most intended to reduce gambling harm may, in part, redirect the most vulnerable players, those seeking to bypass GamStop, towards wholly unprotected environments.Third, the market sustainability question: the near-doubling of Remote Gaming Duty combined with existing compliance cost burdens is driving consolidation and market exit, most visibly in Evoke's prospective forced sale and the departure of GGBet, TGP Europe, and MansionBet. Whilst Entain's CEO frames this as an opportunity for market share gains by larger operators, the concentration of the licensed market into two or three large groups raises longer-term questions about competition, consumer choice, and the pace of innovation in the licensed sector.These findings carry clear policy implications. The most pressing regulatory gap is the absence of a direct affiliate licensing regime. The Gambling Commission's current approach, holding operators responsible for affiliate conduct, is structurally inadequate to address affiliates who promote offshore operators with no UK licence and hence no exposure to UKGC sanction. The House of Lords Select Committee was correct that direct affiliate licensing is necessary to close this gap; the White Paper's decision to reject this recommendation, citing concerns about regulatory burden, should be revisited in light of the evidence that non-GamStop affiliate sites now constitute a significant vector for harmful gambling by self-excluded individuals.The British online casino industry in 2026 is moving towards what may be characterised as 'regulated maturity', a state in which sustainable revenue is derived from a broader base of recreational players within a tightly controlled marketing and product environment (Gambling Commission, 2025a). Whether this transition succeeds will ultimately be judged by two metrics: the extent to which problem gambling prevalence declines as the reforms embed, and the extent to which channelisation is maintained at levels sufficient to deny offshore operators a material foothold. On current evidence, the reform programme is making meaningful progress on the former but remains structurally exposed on the latter, and the affiliate marketing ecosystem sits at the centre of both challenges. References* Sources marked with a star are intentionally left without hyperlinks as these sites are used purely for illustration and comparison of unregulated marketing affiliates.Advertising Standards Authority (ASA) (2025a) Ruling on Dribble Media Ltd t/a Midnite, 17 September 2025. Available at: https://www.asa.org.uk/codes-and-rulings/rulings.html (Accessed: 22 March 2026).Advertising Standards Authority (ASA) (2025b) Ruling on Betway Limited, October 2025. Available at: https://www.asa.org.uk/codes-and-rulings/rulings.html (Accessed: 24 March 2026).Advertising Standards Authority (ASA) (2026) Ruling on Match Bingo, March 2026. Available at: https://www.asa.org.uk/codes-and-rulings/rulings.html (Accessed: 27 March 2026).BestCasinoSites.net (2026a) About us. Available at: https://www.bestcasinosites.net/about-us/ (Accessed: 27 March 2026).BestCasinoSites.net (2026b) Best casino sites - 2026 ratings for UK players. Available at: https://www.bestcasinosites.net/ (Accessed: 25 March 2026).BestCasinoSites.net (2026c) Full list of online casinos with UKGC licences. Available at: https://www.bestcasinosites.net/list/ (Accessed: 27 March 2026).Bigwinboard (2025) The UKGC threatens Bigwinboard: a case of regulatory overreach? Available at: https://www.bigwinboard.com/the-ukgc-threatens-bigwinboard-a-case-of-regulatory-overreach/ (Accessed: 27 March 2026).Bird & Bird (2026) UK gambling reform. Available at: https://www.twobirds.com/en/trending-topics/gambling-reform (Accessed: 27 March 2026).BJ21.com (2025) UK Gambling Commission introduces mandatory bonus wagering calculators. Available at: https://bj21.com/articles/miscellaneous/uk-gambling-commission-introduces-mandatory-bonus-wagering-calculators (Accessed: 27 March 2026).*Bookmakers.bet (2025) Anjouan gaming license: detailed review. Available at: https://www.bookmakers.bet/betting-sites/anjouan/ (Accessed: 27 March 2026).Clifton Davies Consultancy (2025) Affiliates and UKGC face up to licensing uncertainties. Available at: https://cliftondavies.com/affiliates-ukgc-face-licensing-uncertainties/ (Accessed: 27 March 2026).Committee of Advertising Practice (CAP) (2022) CAP Code, Section 16: Gambling. Available at: https://www.asa.org.uk/type/non_broadcast/code_section/16.html (Accessed: 27 March 2026).Deloitte (2025) Gambling duties, TaxScape: Autumn Budget 2025. Available at: https://taxscape.deloitte.com/measures-autumn-budget-2025/gambling-duties.aspx (Accessed: 27 March 2026).Department for Culture, Media and Sport (DCMS) (2023) High stakes: gambling reform for the digital age. London: HMSO. Available at: https://www.gov.uk/government/publications/high-stakes-gambling-reform-for-the-digital-age (Accessed: 27 March 2026).EEGaming (2025) GAMSTOP reports record registrations in 2025, 17 July 2025. Available at: https://eegaming.org/latest-news/2025/07/17/127450/gamstop-reports-record-registrations-in-2025/ (Accessed: 27 March 2026).Entain PLC (2026) 2025 full year results. 5 March 2026. Available at: https://www.entaingroup.com/news-insights/latest-news/2026/2025-full-year-results/ (Accessed: 27 March 2026).Flutter Entertainment (2026) 2025 annual report and accounts. Available at: https://flutter.com/media/le5b5mp4/2025-annual-report-and-accounts.pdf (Accessed: 27 March 2026).Future Market Insights (2026) Online gambling market: global market analysis report, 2035. Available at: https://www.futuremarketinsights.com/reports/online-gambling-market (Accessed: 27 March 2026).Gambling Commission (2024a) Gambling Survey for Great Britain - Annual report (2024): official statistics - Problem Gambling Severity Index. Available at: https://www.gamblingcommission.gov.uk/report/gambling-survey-for-great-britain-annual-report-2024-official-statistics/gsgb-annual-report-2024-problem-gambling-severity-index (Accessed: 27 March 2026).Gambling Commission (2024b) Unlicensed gambling - using data to identify unlicensed operators and estimate the scale of this market, October 2024. Available at: https://www.gamblingcommission.gov.uk/statistics-and-research/publication/unlicensed-gambling-using-data-to-identify-unlicensed-operators-and-estimate (Accessed: 27 March 2026).Gambling Commission (2025a) Business plan and budget 2025 to 2026. Available at: https://www.gamblingcommission.gov.uk/about-us/guide/business-plan-and-budget-2025-to-2026 (Accessed: 27 March 2026).Gambling Commission (2025b) Affiliates or third parties. Available at: https://www.gamblingcommission.gov.uk/licensees-and-businesses/guide/page/affiliates-or-third-parties (Accessed: 27 March 2026).Gambling Commission (2026a) 5.1.1 - Rewards and bonuses - SR code. Available at: https://www.gamblingcommission.gov.uk/licensees-and-businesses/lccp/condition/5-1-1-sr-code (Accessed: 27 March 2026).Gambling Commission (2026b) Gambling promotions to be safer and simpler. Available at: https://www.gamblingcommission.gov.uk/news/article/gambling-promotions-to-be-safer-and-simpler (Accessed: 27 March 2026).Gambling Commission (2026c) Socially responsible incentives: what operators need to know. Available at: https://www.gamblingcommission.gov.uk/blog/post/socially-responsible-incentives-what-operators-need-to-know (Accessed: 27 March 2026).GambleAware (2024) Increase in number of people accessing the National Gambling Support Network. Available at: https://www.gambleaware.org/what-we-do/news/news-articles/increase-in-number-of-people-accessing-the-national-gambling-support-network/ (Accessed: 27 March 2026).GamStop (2025) Our company. Available at: https://www.gamstop.co.uk/corporate (Accessed: 27 March 2026).Global Banking and Finance (2026) Debt-laden William Hill UK owner Evoke explores sale after gambling tax hike. Available at: https://www.globalbankingandfinance.com/evoke-strategic-review-two/ (Accessed: 27 March 2026).Harris Hagan (2026) Marketing archives. Available at: https://www.harrishagan.com/category/marketing/ (Accessed: 27 March 2026).HM Treasury (2025) Gambling duty changes. Available at: https://www.gov.uk/government/publications/changes-to-gambling-duties/gambling-duty-changes (Accessed: 27 March 2026).House of Commons Library (2025) Budget 2025: gambling taxation (CBP-10440). Available at: https://commonslibrary.parliament.uk/research-briefings/cbp-10440/ (Accessed: 27 March 2026).iGaming Business (2025a) Gambling Commission clarifies new bonusing rules to commence in January. Available at: https://igamingbusiness.com/legal-compliance/gambling-commission-clarifies-promotional-rules/ (Accessed: 27 March 2026).iGaming Business (2025b) Gamstop reports surge in self-exclusion among younger bettors. Available at: https://igamingbusiness.com/sustainable-gambling/responsible-gambling/gamstop-surge-self-exclusion-younger-people/ (Accessed: 27 March 2026).iGaming Business (2025c) GamCare warns of rising gambling-related financial issues in the UK. Available at: https://igamingbusiness.com/sustainable-gambling/problem-gambling/gamcare-warns-rising-uk-gambling-related-financial-issues/ (Accessed: 27 March 2026).iGaming Business (2025d) H2GC: UK gambling tax hike will yield only half of Treasury's expected windfall. Available at: https://igamingbusiness.com/finance/tax/h2gc-questions-uk-gambling-tax-hike-yield/ (Accessed: 27 March 2026).iGaming Business (2025e) Can Sweden's Gambling Act update solve all its channelisation problems? Available at: https://igamingbusiness.com/legal-compliance/can-sweden-gambling-act-update-solve-all-its-problems/ (Accessed: 27 March 2026).iGaming Business (2025f) UK sector hit with 40% remote gaming duty, new remote betting tax from 2027. Available at: https://igamingbusiness.com/finance/uk-sector-hit-with-remote-gaming-duty-increase/ (Accessed: 27 March 2026).iGaming Expert (2025a) UKGC in discussions with Meta over illegal gambling ads. Available at: https://igamingexpert.com/features/uk-gambling-black-market-toxic/ (Accessed: 27 March 2026).iGaming Expert (2025b) UKGC: problem gambling stats stable ahead of new controls. Available at: https://igamingexpert.com/features/ukgc-pg-stats-uk/ (Accessed: 27 March 2026).iGaming Today (2026a) UK Gambling Commission ban on mixed-product incentives takes effect 19 January. Available at: https://www.igamingtoday.com/uk-gambling-commission-ban-on-mixed-product-incentives-takes-effect-19-january/ (Accessed: 27 March 2026).iGaming Today (2026b) William Hill owner Evoke weighs sale as UK tax shock and debt pile squeeze options. Available at: https://www.igamingtoday.com/william-hill-owner-evoke-weighs-sale-as-uk-tax-shock-and-debt-pile-squeeze-options/ (Accessed: 27 March 2026).IMARC Group (2026) UK online gambling market size, growth and forecast 2034. Available at: https://www.imarcgroup.com/united-kingdom-online-gambling-market (Accessed: 27 March 2026).Investing.com (2026) Earnings call transcript: Entain PLC H2 2025 shows strong growth, stock rises. Available at: https://www.investing.com/news/transcripts/earnings-call-transcript-entain-plc-h2-2025-shows-strong-growth-stock-rises-93CH-4543337 (Accessed: 27 March 2026).Lancet Public Health (2024) The Lancet Public Health Commission on gambling. The Lancet Public Health, October 2024. Available at: https://www.thelancet.com/journals/lanpub/article/PIIS2468-2667(24)00167-1/fulltext (Accessed: 27 March 2026).Legal 500 (2026) The White Paper: history and status update - country comparative guides. Available at: https://www.legal500.com/guides/hot-topic/the-white-paper-history-and-status-update/ (Accessed: 27 March 2026).Mishcon de Reya (2025) ASA rulings: strong appeal. Available at: https://www.mishcon.com/asa-rulings-strong-appeal (Accessed: 27 March 2026).NEXT.io (2025a) Bigwinboard battles UK Gambling Commission over affiliate legal rights. Available at: https://next.io/news/regulation/bigwinboard-legal-battle-ukgc-affiliate-rights/ (Accessed: 27 March 2026).NEXT.io (2025b) Swedish gaming regulator releases results of black market study. Available at: https://next.io/news/regulation/swedish-gaming-regulator-black-market-study/ (Accessed: 27 March 2026).NEXT.io (2025c) GGBet winding down UK operations after surrendering licences. Available at: https://next.io/news/regulation/ggbet-winding-down-uk-operations/ (Accessed: 27 March 2026).NHS England (2024) NHS tackles problem gambling amid growing demand. Available at: https://www.england.nhs.uk/2024/12/nhs-tackles-problem-gambling-amid-growing-demand/ (Accessed: 27 March 2026).*non-gamstop .uk (2026) Casinos not on GamStop UK. Available at: https://n on-gam stop. uk/ (Accessed: 27 March 2026).Office for Budget Responsibility (OBR) (2025) Betting and gaming duties. Available at: https://obr.uk/forecasts-in-depth/tax-by-tax-spend-by-spend/betting-gaming-duties/ (Accessed: 27 March 2026).Playtech (2025) PAM+. Available at: https://www.playtech.com/products/pam/ (Accessed: 27 March 2026).Professional RakeBack (2025) TGP surrenders UK licences facing £3.3M UKGC fine. Available at: https://professionalrakeback.com/ukgc/tgp-europe-leaves-british-online-gaming-market (Accessed: 27 March 2026).Racing Post (2026) Major changes in the UK iGaming scene in 2026: UKGC update. Available at: https://www.racingpost.com/online-casino/articles/major-changes-in-the-uk-igaming-scene-in-2026/ (Accessed: 27 March 2026).Responsible Affiliates in Gambling (RAiG) (n.d.) RAiG Q&A. Available at: https://www.raig.org/about-us/raig-qa/ (Accessed: 27 March 2026).SBC News (2024) GambleAware: gambling harm on the rise in 2024. Available at: https://sbcnews.co.uk/europe/2024/11/26/gambleaware-gambling-harm-on-the-rise-in-2024/ (Accessed: 27 March 2026).SBC News (2025a) Black market PR stirs up response from UKGC and GAMSTOP. Available at: https://sbcnews.co.uk/europe/uk/2025/05/15/ukgc-gamstop-black-market/ (Accessed: 27 March 2026).SBC News (2025b) Stocks take a hit as Labour confirms gambling tax raises. Available at: https://sbcnews.co.uk/featurednews/2025/11/26/gambling-tax-raises/ (Accessed: 27 March 2026).SBC News (2026a) UKGC ups budget to deliver on key data and evidence projects. Available at: https://sbcnews.co.uk/featurednews/2026/01/08/ukgc-budget-2024-2025/ (Accessed: 27 March 2026).Shoosmiths (2026) 2026 predictions: what's on the horizon for advertising and marketing? Available at: https://www.shoosmiths.com/perspectives/stories/articles/2026-predictions-whats-on-horizon-for-advertising-marketing (Accessed: 27 March 2026).SiGMA (2025a) ASA clarifies gambling ad appeal rules for under-18s. Available at: https://sigma.world/news/asa-clarifies-gambling-ad-appeal-rules-for-under-18s/ (Accessed: 27 March 2026).SiGMA (2025b) Mixed-product incentives and the law of unintended consequences. Available at: https://sigma.world/news/ukgc-mixed-product-incentives-ban/ (Accessed: 27 March 2026).UK Parliament (2018) Gambling advertising: how is it regulated? (CBP-7428). Available at: https://researchbriefings.files.parliament.uk/documents/CBP-7428/CBP-7428.pdf (Accessed: 27 March 2026).United Kingdom Gambling Commission (UKGC) (2017) BGO Entertainment Ltd regulatory settlement, May 2017. Available at: https://www.gamblingcommission.gov.uk/ (Accessed: 27 March 2026).*websitesnotongamstop.uk.net (2026) Non GamStop casinos - best UK sites not on GamStop (2026). Available at: https://websites notongamstop .uk.net/casinos/ (Accessed: 27 March 2026).Wiggin LLP (2025) Gambling promotions face new regulations. Available at: https://wiggin.co.uk/insight/gambling-promotions-face-new-regulations/ (Accessed: 27 March 2026).World Casino Directory (2024) Betting operators surrendering licence could start trend in UK market. Available at: https://news.worldcasinodirectory.com/betting-operators-surrendering-license-could-start-trend-in-uk-market-108833 (Accessed: 27 March 2026).Yogonet (2025a) UK Gambling Commission cracks down on 264 unlicensed operators. Available at: https://www.yogonet.com/international/news/2025/01/23/92802-uk-gambling-commission-cracks-down-on-264-unlicensed-operators (Accessed: 27 March 2026).Yogonet (2025b) Evoke weighs sale or breakup as rising gambling duties deepen financial strain. Available at: https://www.yogonet.com/international/news/2025/12/11/116734-evoke-weighs-sale-or-breakup-as-rising-gambling-duties-deepen-financial-strain (Accessed: 27 March 2026). 

Much of evidence‑based clinical medicine—and the clinical trials that inform it—rests on a classical assumption: that comparative clinical outcomes are determined primarily by biological treatment effects, holding broader social and environmental determinants of health constant. This assumption holds when biological contrasts between treatments are large, but it breaks down in many common clinical decisions in which treatments are biologically similar yet outcomes diverge. Although patient choice has long been acknowledged in such settings, it has been treated primarily as an ethical consideration rather than as a contributing cause of physical outcomes. What has changed is not the presence of patient choice in clinical trials, but what can be identified, interpreted, and acted upon. Across successive generations of trial designs, patient preference has shifted from an implicitly acknowledged feature of care to a measurable contributor to outcomes. Recent reanalysis of the Spine Patient Outcomes Research Trial (SPORT) illustrates this shift by isolating a preference‑mediated component of outcome (β) on the same clinical scale as biologically mediated treatment effects (α). Under near biological equipoise, biological differences were clinically negligible, whereas preference‑mediated effects were large and stable across analyses. Behaviors such as crossover and nonadherence are therefore better understood as structured causal responses rather than as analytic bias. Taken together, these findings define distinct regimes in which outcomes are governed by biology and those in which choice is decisive. Recognizing this distinction constrains both interpretation and action: evidence must identify biological and preference‑mediated effects jointly, decision systems must defer when biology cannot decide, and patients must be supported—through medical literacy—to enact informed choice. Treating patient choice as causal does not weaken evidence‑based medicine; it clarifies its domain and renders preference‑sensitive care coherent and rule‑governed.

AI decision-support systems are increasingly deployed in clinical settings where biological treatment effects are small and outcomes depend materially on patient choice. In prior work, we showed that such preference-sensitive decisions reveal a correctness failure of prediction-centric AI: systems trained to rank options by average biological effects act illegitimately under near equipoise unless they detect preference-sensitive regimes, defer premature ranking, and elicit patient preferences neutrally. That β-aware framework secures non-maleficence by enforcing restraint where biological superiority cannot decide. This paper addresses the downstream question: once safety is assured, what forms of optimization are causally, technically, and ethically admissible? We introduce β-optimization, a constrained framework for AI decision support in preference-sensitive care. Rather than inferring or shaping preferences, β-optimization treats explicitly elicited patient values and feasibility constraints as measured inputs and seeks improvement by maximizing concordance between evidence and what patients value and can sustain. We formalize concordance-based objectives appropriate to observer-dependent decision regimes, specify architectural constraints that preserve neutrality, deferral, and causal separation, and show how large language model–based systems can condition recommendations on elicited preferences without exercising illegitimate authority. We further propose concordance-first evaluation metrics—epistemic, practical, and decisional—for settings where prediction accuracy is ill-posed. Together, β-awareness and β-optimization define a regime-aware theory of AI decision support for preference-sensitive domains: first, do not decide when biology cannot decide; then, once preferences are measured, help decisions succeed—without shaping them.

Large language models (LLMs) increasingly support clinical decisions by synthesizing population-level evidence and estimating average biological treatment effects (α). However, they are not designed to represent the causal contribution of patient choice (β)-a limitation that becomes clinically decisive in preference-sensitive care, where biological differences between options are small. Using the Spine Patient Outcomes Research Trial (SPORT) as an orienting case-a uniquely designed trial in which randomization and patient choice coexisted-we show that preference-mediated effects were clinically meaningful under conditions of near biological equipoise. In such settings, α-centric decision-support systems systematically misrank options because they lack access to β. We formalize this limitation as "α-bias" and "β-blindness," and propose a regime-routed architecture-Detect → Elicit → Recommend → Learn-with explicit deferral rules, neutrality constraints, provenance, and auditable guardrails. When outcomes hinge on choice rather than biological superiority, elicitation is a precondition for recommendation, not an optional refinement. These principles extend beyond medicine to any domain in which outcomes depend on decision-contingent preferences rather than fixed parameters. When outcomes depend on choice rather than biology, improving choices is improving outcomes. A companion paper addresses how β-aware systems may permissibly improve concordance once these safety conditions are satisfied and thus AI decision support moves to βoptimization.

We present HAR-Agent, a multilingual multimodal human activity recognition system that unifies visual, audio, and text perception through a common textual representation, with a large language model serving as the central decision-maker. We build a multimodal agent architecture and systematically compare 17 model configurations-varying parameter count from 1.5 billion to 72 billion and training method between instruction tuning and supervised fine-tuning-to determine which produces the most effective activity recognition agent. Instruction tuning outperforms supervised fine-tuning by 17 to 28 percentage points at every scale, and a striking inverse scaling phenomenon emerges whereby the smallest 1.5-billionparameter student achieves the highest student accuracy at 41.4%, outperforming all larger students whilst requiring under one gigabyte of GPU memory. Knowledge distillation from a 72-billion-parameter teacher achieves moderate agreement at 48 times compression. A multilingual audio benchmark across five languages yields 89.2% accuracy with sub-second latency, and cross-domain evaluation on the Toyota Smarthome dataset confirms that instruction-tuned agents generalise better to unseen real-home environments. These findings provide practical guidelines for building and deploying large language model based activity recognition agents on consumer hardware.

This book presents a sustained philosophical and critical inquiry into Sudanese art under conditions of war, displacement, and historical fragmentation, advancing Post-Interpretive Criticism (PIC) as an ethical alternative to dominant interpretive frameworks in contemporary art discourse. Rather than treating artworks as objects to be decoded, explained, or culturally translated for external audiences, this study repositions the critic as a witness in proximity—one who resists extraction, narrative closure, and symbolic appropriation.Through a series of close encounters with Sudanese visual practices—ranging from material craft traditions and domestic forms to contemporary expressions shaped by exile and rupture—the book develops a methodology grounded in restraint, presence, and moral adjacency. It argues that conventional art criticism, particularly within post-1950s interpretive paradigms, has produced a condition of interpretive saturation, wherein artworks are reduced to vehicles for meaning rather than approached as sites of lived continuity and fragile coherence.Drawing on the philosophical architecture of Post-Interpretive Criticism—including concepts such as viewer-as-evidence, ethical proximity, and the refusal of interpretive closure—the text examines how Sudanese art operates not as representation but as survival infrastructure: a means of carrying memory, dignity, and cultural continuity through conditions of systemic disruption. In doing so, it reframes aesthetic experience as an ethical encounter shaped by limits, rather than an opportunity for discursive expansion.The book further situates Sudanese art within broader discussions of archive politics, institutional mediation, and the epistemic violence of interpretation, arguing that acts of naming, framing, and explaining often function as forms of displacement. Against this, it proposes a discipline of witnessing that privileges silence, material attention, and positional awareness over interpretive authority.Positioned at the intersection of aesthetics, anthropology, and philosophy, Threads of Sand, Scar, & Scripture contributes to emerging post-hermeneutic approaches to art criticism by offering both a theoretical framework and a series of applied analyses. It establishes Post-Interpretive Criticism not merely as a method, but as a necessary ethical recalibration in the face of cultural fragility, where the role of criticism is no longer to extract meaning, but to remain answerable to what survives.Dorian Vale is a chosen pseudonym, not to obscure identity, but to preserve clarity of voice and integrity of message. It creates distance between the writer and the work, allowing the philosophy to stand unclouded by biography. The name exists not to hide, but to honor the seriousness of the task: to speak without spectacle, and to build without needing to be seen.Post-Interpretive Criticism, Stillmark Theory, Message-Transfer Theory, MTT, Misplacement, Displacement, Aesthetic Displacement Theory, Theory of Misplacement, Absential Aesthetics, Witness Aesthetics, Hauntmark Theory, Spiritual Criticism, Presence-Based Criticism, Custodianship of Art, Art as Ontology, Aesthetic Recursion Theory, Aesthetic Recursion, Viewer as Evidence Theory, Restraint in front of art, Moral proximity, Interpretive silence, Erasure as ethics, Temporal scarcity, Silence as method, Ontology of beauty, Aesthetic mercy, Language as violence, Art encounter ethics, Epistemology of witness, Philosophy of Art, Aesthetics, Art Theory, Contemporary Aesthetics, Comparative Aesthetics, Phenomenology and Art, Ethics in Art Criticism, Interpretation and Meaning, Criticism and Reception Theory, Epistemology of Art, Visual Culture Studies, Dorian Vale, Founder of Post-Interpretive Criticism, Post-Aesthetic Critic, Independent Philosopher of Art, Museum of One, Art Writer and Theorist, Aesthetic Philosopher, Custodian of Witness Aesthetics, Spiritual Aesthetics Movement, The Doctrine of Post-Interpretive Criticism, The Custodian’s Oath, The Canon of Witnesses, Art as Truth, Art as Presence, The Viewer as Evidence, Interpretation vs. Witnessing, Language as Custody, Erasure as Afterlife, Museum of One Manifesto, Alternative art criticism, New art criticism movement, Ethical art theory, Criticism beyond interpretation, Slow looking philosophy, Quiet philosophy of art, Radical art restraint, Witness over interpretation, Interpretive Restraint, The Journal of Post-Interpretive criticism, The Journal of Post-Interpretive criticism ISSN 2819-7232), The Journal of Post-Interpretive Criticism (Q136530009), Epoché Fidelity Index (EFI) (Q138018710), Phenomenological Phase Alignment Score (PPAS) (Q138018807), Residue Engagement Restraint Ratio (RERR) (Q138018901), Quasi-Subject Agency Recognition Index (QSARI) (Q138018929), Dialectical Circulation Index (DCI) (Q138018950)

A document by J. Roberto Sánchez-Reina. Click on the document to view its contents.

We propose a novel mechanism for the stabilization of neutron stars beyond the classical Tolman-Oppenheimer-Volkoff (TOV) limit. By introducing an emergent "Geometric Pressure" (Pent) derived from quantum entanglement, we show that the transition from area-law to volume-law entropy scaling prevents gravitational singularities. Our numerical results indicate that stable configurations exist for M > 2.5M⊙, providing a theoretical basis for recent observational anomalies such as the secondary component of GW190814.

The finalization of NIST FIPS 203 (ML-KEM), FIPS 204 (ML-DSA), and FIPS 205 (SLH-DSA) in August 2024 marked the beginning of the global post-quantum cryptography (PQC) migration. However, most existing PQC implementations target a single programming language, require deep cryptographic expertise, and lack the hybrid constructions and protocol integration needed for real-world deployment. This paper presents an open-source project implementing all three NIST PQC standards-plus hybrid key encapsulation mechanisms (X25519+ML-KEM-768, ECDH-P256+ML-KEM-768, and higher-security variants), composite digital signatures (ML-DSA+Ed25519, ML-DSA+ECDSA-P256), and TLS 1.3 integration building blocks with named groups, signature algorithms, and cipher suites-across eight programming languages: Rust, Go, JavaScript/ESM, Python 3.12+, Java 17+, C#/.NET 10, Swift 5.9+, and PHP 8.1+. To the best of our knowledge, no other single open-source project provides this combination of breadth across all three standards, eight languages, hybrid constructions, and TLS protocol integration, although several excellent libraries cover subsets of this scope. We present the mathematical foundations for each algorithm: the Module Learning With Errors (M-LWE) problem and its hardness reduction, the Number Theoretic Transform (NTT) over the cyclotomic ring R q = Z q [X]/(X 256 + 1) with q = 3329, the Fujisaki-Okamoto transform for IND-CCA2 security in the quantum random oracle model (QROM), the Fiat-Shamir with aborts paradigm for lattice signatures, and the hypertree construction for stateless hash-based signatures. We document security hardening techniques (branchless field arithmetic, constant-time ciphertext comparison, heap-free SHAKE-256 streaming, zero-allocation Keccak), performance optimizations (Barrett reduction with full 3329 2-pair correctness testing, compile-time NTT zeta tables, ThinLTO profiles), and a cross-language interoperability test suite. The hybrid key exchange framework implements a SHA3-256 domain-separated KDF combiner construction following the framework of Bindel et al. [9]. The TLS 1.3 integration provides a three-step handshake API (GenerateKeyShare/CompleteKeyExchange/RecoverSharedSecret) that enables developers to add PQC building blocks to existing TLS stacks. All implementations pass NIST Known Answer Test (KAT) vectors for every supported parameter set. Cross-language interoperability has been validated across all 12 SHAKE-based parameter sets (3 ML-KEM, 3 ML-DSA, 6 SLH-DSA): Python generates test vectors and all seven other languages (Go, Java, JavaScript, Rust, Swift, C#/.NET, PHP) independently verify them-96/96 tests passing with zero failures. The complete source code is available under the MIT license at https://github.com/liviuepure/PQC-Standards-Implementation.

Seepage springs are a common freshwater habitat in the Washington D.C. area that are home to a variety of invertebrates including the endemic and rare Stygobromus hayi, the official amphipod of Washington DC, USA, but little is known about the community structure of these habitats. We analyzed the general food web dynamics of two seepage springs (Pimmit Run and Goldmine Tract) in the Washington D.C. area that are known to include a congener of S. hayi, Stygobromus tenuis potomacus. Within these two seepage springs, we identified 29 taxa, with varying degrees of abundances, from 6 field excursions over three seasons: Winter, Spring, and Summer. Using dual abundance stable isotope analysis of δ13C and δ15N, we discerned the trophic positions of Lumbriculidae, Conasellus kenki, S. tenuis potomacus, and Crangonyx shoemakeri, and estimated the trophic positions of Tipula, Pseudolimnophila, and a species of Platyhelminthes. S. tenuis potomacus was the dominant predator in both locations with δ15N and δ13C values between 7 and 10‰ and -25 and -27‰ respectively. C. kenki and C. shoemakeri were consistently lower in the food web, with δ15N and δ13C values between 4 and 7‰ and -24 and -27‰ respectively and, in some cases, C. kenki maybe prey for S. tenuis potomacus. In both seeps leaf material was 4 to 5‰ lower in δ13C than the shredding/grazing invertebrates suggesting that they derive nutrients from microbes colonizing leaves not from leaf carbon itself.

Single-atom catalysts based on iridium have attracted significant attention for water splitting. However, the role of Ir oxidation states and their dependence on the electronic structure of the support material are not well understood. Here, we investigate the catalytic behavior of single-atom Ir anchored on anatase TiO2 nanofilms, using this platform system to study the valence-dependent activity toward hydrogen evolution reaction (HER) and oxygen evolution reaction (OER). High-valence Ir6+ single atoms exhibit excellent HER activity and outstanding operational stability under alkaline conditions, including at elevated electrolyte temperatures at 80 °C, but show negligible OER activity on TiO2 nanofilms. Through controlled electrochemical cyclic voltammetry and secondary annealing, Ir6+ species are partially converted into Ir4+ along with a concurrent reduction in deep-trapped oxygen vacancies and reconstruction of the local electronic environment of the host TiO2. This valence modulation leads to a pronounced enhancement in OER performance, with Ir4+ species dominated single-atom sites delivering lower overpotentials with –0.42 ± 0.04 V vs RHE at 10 mA cm⁻2. Density functional theory calculations reveal more favorable adsorption energetics for OER intermediates on Ir4+ compared to Ir6+ sites, which is in agreement with our experimental observations.