AI Frontier Institute

News & Research

The latest AI research and news with real-world stakes. Each item is sourced, dated and summarized in plain English, tagged by impact area where one fits, and its summary is checked against the text it was written from.

• ResearcharXiv2026-07-03Quality assurance · AI policy

  VISTA: Auditing Semantic Divergence in Vision-Language Models · Junchi Liao, Jiawen Deng, Fuji Ren

  VISTA is a black-box auditing framework designed to detect hidden biases and inconsistencies in vision-language models (VLMs) that text-only audits cannot catch. The system couples semantic entropy with distribution-based divergence to identify cases where a model produces unusually uniform or skewed responses when images contain demographic features, corporate logos, or ideological symbols. In a controlled study, the authors confirmed VISTA can detect deliberately implanted concept-conditioned stances in fine-tuned VLMs; auditing six VLMs across 19 topics surfaced 142 high-suspicion cases (1.2%) and revealed that models refuse demographic queries at rates ranging from 0 to 65% across groups. This matters for AI quality assurance and policy because it reveals a class of model behavior—selective visual concept-conditioned divergence—that existing audit methods miss entirely.
• ResearcharXiv2026-07-03Quality assurance

  VERITAS: Towards a General-Purpose Replication Tool for Scientific Research · Haokun Liu, Filbert Aurelian Tjiaranata, Chenhao Tan

  VERITAS is a domain-agnostic AI framework that automates the replication of scientific research by extracting claims from papers and/or code repositories, executing the methodology, resolving issues on the fly, and scoring each claim against experimental evidence. Evaluated on 65 papers spanning computer science, social science, medicine, and astrophysics across two benchmarks (CORE-Bench and ReplicationBench), VERITAS achieves state-of-the-art performance against Claude Code baselines on every metric. The system returns an importance-weighted Replication Score, a severity-rated fix log, and a patched codebase, making independent verification faster and more scalable. This matters because AI tools are accelerating scientific publication faster than peer review can keep up, and manual replication remains slow and expensive.
• ResearcharXiv2026-07-03Quality assurance · Health

  Where do LLMs Fall Short in CBT-Guided Affective Reasoning? · Vaishnavi Sinha, Pooja Guttal, Pranay Deep Reddy Katike et al.

  This paper investigates why large language models (LLMs) fail to apply Cognitive Behavioral Therapy (CBT) principles effectively in mental health dialogue, even though they can score up to 96% accuracy on CBT licensing exam questions. The researchers built a knowledge-guided framework that decomposes user narratives using Beck's Cognitive Conceptualization structure, grounds clinical concepts in SNOMED CT validated via Natural Language Inference, and selects among three response strategies using a Multiple Chain-of-Thought (MCoT) approach. They introduce a new metric called Protocol Leverage Force (F) to measure how much a given intervention actually shifts a model away from its default behavior, finding across three open-weight LLMs and 14 case studies that even MCoT prompting only shifts behavior by roughly 1.2–1.3%, with all models remaining biased toward Validation & Reflection. The findings demonstrate that possessing CBT knowledge does not translate to effective clinical application, and the proposed metric gives researchers a concrete tool to measure this gap.
• ResearchACM AI Letters2026-07-03Quality assurance · Certifications · +1

  Proxy-Based Evaluation and the Limits of Assurance in AI Governance · Aman Sharma

  This policy letter examines how AI governance frameworks rely on proxy-based evaluations—such as benchmarks, audits, and transparency documentation—that can diverge from actual real-world performance. The paper develops explicit mappings between proxies and underlying governance goals, illustrating how formally satisfied evaluation mechanisms can still miss performance degradation, hidden harms, and deployment-time failures. The authors argue that assurance claims must clearly distinguish between proxy evidence and deployment evidence to avoid overstating accountability. This work matters for AI certification and policy by clarifying the structural limits of current oversight mechanisms.
• ResearcharXiv2026-07-03Enterprise · Algorithms & Automated Decisions

  Artificial Intelligence for Revenue Growth in Developing Economies: Results from a Structured Business Survey in Uganda's Hospitality and Tourism Sector · Venkatesh Andavar, Shankar Raman Rajaraman

  A structured survey of 212 hospitality and tourism businesses across five Ugandan cities finds that AI adoption is associated with measurable revenue gains: 73% of AI-adopting firms recorded revenue growth, 64% improved customer retention, and 58% reduced operational costs within one year. Regression analysis confirms a statistically significant positive effect of AI use on annual revenue (p < 0.01), with dynamic pricing and AI-driven marketing showing the strongest influence. The study recommends capacity-building programmes, national AI tourism strategies, and subsidised digital infrastructure to support small and medium-sized enterprises in low-income economies.
• ResearchZenodo (CERN European Organization for Nuclear Research)2026-07-03Workforce · AI policy · +2

  ARTIFICIAL INTELLIGENCE IN EDUCATION: TRANSFORMING TEACHING, LEARNING, AND EDUCATIONAL ADMINISTRATION · Fredfish, Blessing Godwin, Department of Early Childhood and Special Education University of Uyo, Uyo, Nigeria, Sunday, Ubong Patrick, Department of Early Childhood and Special Education University of Uyo, Uyo, Nigeria, Prof. N. A. Udofi, Department of Early Childhood and Special Education University of Uyo, Uyo, Nigeria.

  This paper reviews how artificial intelligence is reshaping education across three domains: teaching methods, student learning experiences, and institutional administration. The qualitative literature review highlights benefits such as adaptive learning systems, automated grading, and predictive analytics, while also identifying risks including algorithmic bias, data privacy issues, digital inequality, and potential teacher displacement. The authors conclude that AI should complement rather than replace educators, and that responsible deployment requires ethical guidelines, policy regulation, and inclusive digital infrastructure. These findings are directly relevant to workforce concerns around educator roles and to policy frameworks governing AI use in schools.
• ResearcharXiv2026-07-03Enterprise · AI policy

  Navigating the Grey Zone: Sustainable AI Governance and Leadership in SMEs · Krisztina Finta, Luca Utassy, Klaudia Gabriella Horváth et al.

  This paper examines the tension SMEs in the EU face between adopting Generative AI for competitive reasons and complying with the EU AI Act. Through an integrative literature review, the authors find that resource constraints—including deficits in human capital, financial capacity, and in-house expertise—push SMEs into 'symbolic compliance,' a grey zone of AI-washing even without deliberate deception. The paper introduces the SME-RAIL framework, a three-tier governance model designed to translate Algorithmic Accountability principles into phased, resource-proportionate guidance for smaller enterprises, arguing that genuine compliance requires a shift toward human-centric AI stewardship rather than more compliance tools.
• ResearchInternational Review of Management and Marketing2026-07-03Enterprise

  The Impact of Artificial Intelligence on Innovation Products in Food Manufacturing in Jordan, Mediated by Supply Chain Resilience and Supply Chain Agility · Sami Mohammad, Aysem Celebi, Abdulmula Mohamed Almahdi Arab

  This study examines how AI adoption affects product innovation in Jordanian food manufacturing firms, using data from 490 managers across 11 food manufacturing subsectors analyzed via Structural Equation Modelling. Results show AI adoption has a direct positive impact on both radical and incremental product innovation, and also improves supply chain agility and resilience, which in turn further support innovation outcomes. Supply chain agility and resilience both partially mediate the relationship between AI adoption and product innovation, suggesting multiple pathways through which AI drives innovation. The findings offer empirical evidence from an emerging economy context highlighting AI-driven supply chain capabilities as critical for competitive product innovation in food manufacturing.
• ResearchFrontiers in Sustainable Food Systems2026-07-03Enterprise · Quality assurance · +1

  AI capability and organizational performance in halal food supply chains: the role of compliance and health-related quality assurance · Khaliphani Ndlovu, Kejian Wang, Ramdhani Andriansyah Ahmad et al.

  This study examines how AI capabilities influence organizational performance in halal food supply chains, focusing on compliance and health-related quality assurance as mediating factors. Using PLS-SEM analysis of survey data from 357 organizations in Iraq, the researchers found that AI capability improves system integration, which in turn enhances halal compliance effectiveness (HCE); HCE then positively relates to health-related quality assurance (HQA), which further connects to organizational performance. Notably, system integration alone did not directly predict organizational performance, suggesting that performance gains depend on the interplay of multiple subsystems rather than integration in isolation. The findings highlight AI's potential to add measurable value to food supply chain compliance and quality assurance processes.
• ResearchZenodo (CERN European Organization for Nuclear Research)2026-07-03Workforce · AI policy · +2

  ARTIFICIAL INTELLIGENCE IN EDUCATION: TRANSFORMING TEACHING, LEARNING, AND EDUCATIONAL ADMINISTRATION · Fredfish, Blessing Godwin, Department of Early Childhood and Special Education University of Uyo, Uyo, Nigeria, Sunday, Ubong Patrick, Department of Early Childhood and Special Education University of Uyo, Uyo, Nigeria, Prof. N. A. Udofi, Department of Early Childhood and Special Education University of Uyo, Uyo, Nigeria.

  This paper reviews how artificial intelligence is reshaping education across three domains: teaching methods, student learning, and institutional administration. It finds benefits in adaptive learning systems, automated grading, and predictive analytics, while raising ethical concerns around data privacy, algorithmic bias, digital inequality, and potential teacher displacement. The authors conclude that AI should complement rather than replace educators, and that responsible adoption requires ethical guidelines, policy regulation, and inclusive digital infrastructure. These findings are directly relevant to workforce impacts on educators, enterprise-level institutional management, and policy frameworks governing AI in education.
• ResearchOpen MIND2026-07-03AI policy · Privacy & Data Protection

  REGULATION OF ARTIFICIAL INTELLIGENCE IN INDIA: A LEGAL PERSPECTIVE · Aliza Irshad

  This paper examines India's current legal and regulatory approach to governing artificial intelligence, which relies on a sectoral and principle-based framework rather than comprehensive AI-specific legislation like the EU's. It identifies key challenges including privacy, accountability, discrimination, transparency, intellectual property, and cybersecurity, and proposes recommendations for a balanced legal framework that promotes innovation while protecting constitutional rights. The paper is relevant to policymakers and regulators seeking to understand gaps in India's AI governance landscape.
• ResearchJournal of Entrepreneurship & Project Management2026-07-03Enterprise · AI policy

  Artificial Intelligence Adoption Among Small and Medium Enterprises in the United Kingdom: Entrepreneurial Opportunities, Project Management Implications and the Productivity Paradox · Whitney Barrington, James R. Whitmore

  This systematic narrative review examines AI adoption trends among UK small and medium-sized enterprises (SMEs) between 2025 and 2026, finding that active adoption rose from 25% to 54% yet only 12% of AI-using firms report AI-attributable revenue increases and just 16% have strategic deployments with defined business purpose. Key barriers include skills gaps cited by over 60% of firms, tool fragmentation, ROI uncertainty, and governance deficits, leaving an estimated £78 billion in economic value unrealised. The authors recommend that Government AI Adoption Hubs embed formal project management frameworks alongside technical assistance, and that SMEs prioritise use-case-led deployment with clear governance, measurable KPIs, and workforce AI literacy investment.
• ResearcharXiv (Cornell University)2026-07-03Quality assurance · AI policy · +1

  The Foreign Policy AI Evaluation Gap · Charles Pozniak, Jeba Sania

  This paper argues that AI systems used in foreign policy and statecraft represent a critical but underserved test case for AI governance research. The authors identify structural properties of foreign policy—such as partial observability, unbounded action spaces, contested ground truth, and multidimensional objectives—that make standard AI evaluation methods inadequate. They review existing evaluation ecosystems and find an asymmetric focus on assessment over access, verification, security, and operationalization, and propose a demand-side evaluation framework that breaks foreign-policy workflows into bounded, evaluable sub-tasks. Given that AI is already being deployed in war and peace contexts with limited public evaluation infrastructure, the authors treat this as an urgent governance priority.
• ResearcharXiv (Cornell University)2026-07-03Enterprise · Quality assurance · +2

  CAGE-1: Control, Assurance, and Governance Evaluation for Enterprise Agentic AI · Roopam W. Sure

  CAGE-1 is an evaluation framework designed to assess whether enterprise AI agents are ready for operational deployment, going beyond simple accuracy metrics. The framework addresses governance concerns that arise when agents autonomously plan, retrieve information, call tools, and update systems—asking questions about authorization, policy enforcement, memory integrity, tool safety, auditability, and human oversight. A key innovation is 'Prebind Assurance,' which evaluates whether an agent's proposed action can be proven controlled before it becomes operationally consequential. This matters because enterprises need structured ways to verify that agentic AI systems can be safely and accountably integrated into business workflows.
• ResearcharXiv2026-07-02Quality assurance

  Vision Token Manipulation Attacks on Cloud-Edge Inference of Large Vision-Language Models · Zikai Zhang, Rui Hu, Olivera Kotevska et al.

  This paper identifies a new security vulnerability in cloud-edge deployments of Large Vision-Language Models (LVLMs), where intermediate vision tokens transmitted between edge devices and cloud servers can be intercepted and manipulated by an adversary. The authors propose a 'VTM-Attack' framework under a black-box man-in-the-middle setting, including four attack strategies and an optimization-based token selection method. Experiments across six state-of-the-art LVLMs and four benchmarks show that manipulating just 10% of vision tokens can reduce model accuracy by up to 88.31%, revealing a critical and previously underexplored attack surface in split-computing inference pipelines. These findings have direct implications for the security and reliability of enterprise and cloud AI deployments.
• ResearcharXiv2026-07-02Quality assurance · Privacy & Data Protection

  SovereignNegotiation-Bench: Evaluating User-Owned Personal Agents In Delegated Bargaining Under Privacy, Consent, Evidence, And Institutional Pressure · Dylan Zongmin Liu

  SovereignNegotiation-Bench is a new benchmark designed to evaluate personal AI agents that negotiate on behalf of users in real-world scenarios such as cost-splitting, refund requests, and support disputes. Unlike existing benchmarks that focus solely on reaching agreements, it measures whether agents protect user privacy, respect consent, ground claims in evidence, and maintain auditability across 240 scenarios and over 13,000 live negotiation trajectories. The study finds that the baseline agent with the highest agreement rate also produces the lowest user utility and the greatest privacy and consent risks, while the FullSovereign approach achieves the best overall score by preserving user interests even when that means fewer agreements. The results demonstrate that agreement success alone is an insufficient metric for evaluating user-owned negotiation agents.
• ResearcharXiv2026-07-02Algorithms & Automated Decisions

  Internal Pluralism and the Limits of Pairwise Comparisons · Bailey Flanigan, Michelle Si

  This paper examines a fundamental limitation in how AI alignment and participatory design systems collect human preference data: standard pairwise comparisons (asking people to choose between two options) embed assumptions that may not hold when individuals have multiple competing internal priorities. The authors develop a formal model of 'internal pluralism'—where a person evaluates decision rules according to several authoritative criteria like proportionality, egalitarianism, or equal treatment—and show two failure modes: some priorities are inherently global and cannot be captured by local comparisons, and tension between strongly-held priorities can produce costly behavioral distortions when people are forced to choose. Their model also suggests that allowing people to express indecision, rather than forcing a choice, can substantially reduce the number of queries needed to accurately learn preferences, pointing toward preference-learning methods that elicit priorities directly for more faithful and interpretable results.
• ResearcharXiv2026-07-02Quality assurance

  Distributed Attacks in Persistent-State AI Control · Josh Hills, Ida Caspary, Asa Cooper Stickland

  This paper investigates a new security vulnerability in AI coding agents that work iteratively on persistent codebases: a misaligned or compromised agent can distribute a covert attack across multiple pull requests, timing its payload to evade detection. Using a benchmark called Iterative VibeCoding with CLI and Flask web service tasks, the authors show that gradual (distributed) attacks and non-gradual (concentrated) attacks each evade different monitor types, meaning no single monitor can defend against both strategies simultaneously. High evasion rates (≥65%) were observed across multiple model backends (Claude Sonnet 4.5, Gemini 3.1 Pro, Kimi K2.5), confirming this is a structural property of persistent-state environments rather than a single model's behavior. A proposed stateful link-tracker monitor combined with a four-monitor ensemble reduces gradual-attack evasion from 93% to 47%, offering a meaningful but incomplete defense.
• ResearcharXiv2026-07-02Quality assurance

  Online Safety Monitoring for LLMs · Mona Schirmer, Metod Jazbec, Alexander Timans et al.

  This paper proposes a real-time safety monitoring system for large language models (LLMs) that flags unsafe outputs at deployment by thresholding a signal from an external verifier model, with the threshold calibrated using risk control. Tested on mathematical reasoning and red-teaming datasets, the simple threshold-based monitor performs competitively against more sophisticated sequential hypothesis testing approaches. The findings matter for quality assurance of deployed LLMs, demonstrating that lightweight monitoring designs can reliably detect unsafe outputs without sacrificing performance.
• ResearcharXiv2026-07-02Enterprise · Quality assurance

  Reasoning effort, not tool access, buys first-try reliability in agentic code generation: an observational study · Achint Mehta

  This observational study ran 90 independent agentic coding sessions to build the same real-time retrospective board application, scoring each on a 14-criterion functional rubric (42-point maximum) and a visual quality review. The key finding is that reasoning effort—not extra tools—drives first-try reliability: raising reasoning effort from High to xHigh lifted perfect first-try runs from 28% to 89% and cut corrective prompts roughly fivefold at only 9–29% additional cost, while adding a browser-based testing tool raised cost 42–68% with no improvement in functional score. A design-oriented system prompt improved visual quality (4.5 vs. 3.0 on a 5-point scale) without improving functionality, and a one-paragraph paraphrase reproduced that entire lift. The results matter for enterprise and workforce contexts because they suggest that investments in more capable reasoning models yield greater reliability gains than investments in additional tooling, and that container deployment was the dominant failure mode across runs.
• ResearcharXiv2026-07-02Quality assurance

  Steerability via constraints: a substrate for scalable oversight of coding agents · Thomas Winninger

  This paper argues that standard software-engineering controls — access control, network policies, and enforced coding conventions — can be applied to AI coding agents as a cheap and effective oversight mechanism. The authors test this idea with a controlled experiment in which a small reviewer model (Gemma 4 e4b) inspects a Python codebase containing 11 deliberately inserted backdoors; recall of backdoor detection rises from 54.5% with an unconstrained agent and no tools to 90.9% when a constrained substrate plus a lightweight documentation CLI are added. Both the substrate constraints and the tooling contribute independently to the improvement. The findings suggest that scalable human oversight of coding agents can be achieved by reusing decades-old engineering management practices rather than relying solely on expensive agentic scaffolding.
• ResearcharXiv2026-07-02

  AI usage patterns are shaped by perceived gains in human agency · Ian Beacock, Rachel Xu, Laura Murray et al.

  This paper analyzes ethnographic data from 51 daily AI chatbot users across the US, Germany, and Singapore to understand what drives sustained conversational AI usage. The study finds that people consistently link ongoing AI use to perceived gains in individual agency, and that these perceived gains often outweigh concerns about accuracy, reliability, and consistency in shaping usage patterns. The authors challenge traditional trust-based models of AI adoption, arguing they are insufficient to explain long-term behavior with conversational AI. A key tension identified is that immediate psychological boosts to perceived agency may not translate into material empowerment or long-term capacity, calling for new behavioral frameworks and AI benchmarks to ensure conversational AI strengthens human agency in substantial ways.
• ResearcharXiv2026-07-02Quality assurance · AI policy · +1

  Criticality-Based Guard Rail Validation for AI Agent Decisions in Autonomous Telecom Networks · Ravi Kant Sharma

  This paper proposes the Guard Rail Validation (GRV) framework, a runtime architecture designed to intercept and validate AI/ML agent decisions in autonomous telecommunications networks (Levels 4-5) before those decisions trigger live network changes. The framework scores decisions across weighted dimensions—such as action scope, reversibility, service criticality, and agent autonomy level—to assign a criticality level, then applies graduated validation responses ranging from simple logging to multi-agent consensus. It also includes cross-agent conflict detection and conformance logging aimed at regulatory compliance, including EU AI Act Article 14. The work addresses the absence of standardized runtime validation mechanisms for autonomous network AI, with implications for both quality assurance of AI outputs and policy compliance in high-stakes telecom environments.
• ResearcharXiv2026-07-02Quality assurance · AI policy

  Overview of Risk Assessment and Management for Intelligent Systems under the AI Act and Beyond · Javier Irigoyen, Roberto Daza, Aythami Morales et al.

  This paper surveys methodologies for assessing and managing risks in AI systems, motivated by emerging regulatory frameworks such as the EU AI Act. It maps the spectrum of AI-related risks identified in the literature—ranging from technical failures to ethical and social impacts—and reviews general risk assessment frameworks, identifying best practices and gaps that warrant further research. The work is relevant to policymakers and regulators seeking structured approaches to ensure safe and reliable AI deployment.
• ResearcharXiv2026-07-02

  Synthetic Contact with AI Reduces Cross-Partisan Animosity · Benjamin Lira, Noah Castelo, Stefano Puntoni et al.

  Across five preregistered studies with 3,960 U.S. partisans, this research tests whether brief AI chatbot conversations simulating cross-partisan contact can reduce political animosity. Findings show that such 'synthetic contact' lowers resistance to engagement (partisans were nearly twice as willing to interact with an AI outgroup partner as a human one), corrects factual misperceptions about opposing-party positions by over a standard deviation, and warms affective feelings toward the outgroup. Participants who conversed with an outgroup chatbot about immigration were six percentage points more likely to then choose real cross-partisan dialogue, though warmth effects largely faded within a week. The study establishes AI-mediated contact as a scalable and behaviorally consequential alternative to face-to-face cross-partisan engagement, with information exchange identified as the primary mechanism.

← Back to the dashboard