arXiv subjects

Marc-Oliver Pahl

Publications and source records attributed to Marc-Oliver Pahl.

15 recordsLinked to original sources

arXiv2026

Experts Disagree on How to Fight AI Disinformation, but Agree That Health and Politics Need Different Solutions

When 54 international experts assessed AI-generated disinformation threats, they revealed a surprising pattern: while video deepfakes received the highest average threat ratings in the political domain (M = 6.31/7), the pattern differed in the health domain, where AI-generated text received the highest average rating (M = 5.80). Experts also diverge on what to do: government regulation drew both the most "most effective" (30%) and the most "least effective" (15%) votes, though rating distributions were contested rather than polarized, indicating disagreement over priorities rather than over efficacy. These findings offer an initial expert map of an AI-disinformation landscape that is still rapidly forming.

cs.CY↗

arXiv2026

Interrupting the Chain: Human Perception of AI-Generated Disinformation Through a Kill Chain Lens

Generative AI enables customized misinformation at scale, yet defenses remain largely reactive. We present empirical findings from a human-subject study (n=504 participants, n=2,438 judgments) in which users classified news fragments by origin (human vs. machine) and veracity (real vs. fake). We organize results using an adapted cybersecurity kill chain as a taxonomy for intervention, mapping perception data onto stages of a cognitive attack lifecycle. Three key findings emerge: (1) a perception-accuracy gap where heightened suspicion does not improve detection; (2) modern LLMs frequently produce human-indistinguishable text; and (3) an asymmetric cognitive fatigue effect where fake-news detection degrades by 10.2 percentage points under sustained exposure while AI-origin detection remains stable. These findings identify candidate intervention points for proactive defense against AI-driven disinformation.

cs.CY↗

arXiv2026

Can Humans Tell? A Dual-Axis Study of Human Perception of LLM-Generated News

Can humans tell whether a news article was written by a person or a large language model (LLM)? We investigate this question using JudgeGPT, a study platform that independently measures source attribution (human vs. machine) and authenticity judgment (legitimate vs. fake) on continuous scales. From 2,318 judgments collected from 1,054 participants across content generated by six LLMs, we report five findings: (1) participants cannot reliably distinguish machine-generated from human-written text (p > .05, Welch's t-test); (2) this inability holds across all tested models, including open-weight models with as few as 7B parameters; (3) self-reported domain expertise predicts judgment accuracy (r = .35, p < .001) whereas political orientation does not (r = -.10, n.s.); (4) clustering reveals distinct response strategies ("Skeptics" vs. "Believers"); and (5) accuracy degrades after approximately 30 sequential evaluations due to cognitive fatigue. The answer, in short, is no: humans cannot reliably tell. These results indicate that user-side detection is not a viable defense and motivate system-level countermeasures such as cryptographic content provenance.

cs.CY↗

arXiv2026

CRED-1: An Open Multi-Signal Domain Credibility Dataset for Automated Pre-Bunking of Online Misinformation

This article presents CRED-1, an open, reproducible domain-level credibility dataset combining two openly-licensed source lists (OpenSources.co and Iffy.news) with four computed enrichment signals: domain age (WHOIS/RDAP), web popularity (Tranco Top-1M), fact-check frequency (Google Fact Check Tools API), and threat intelligence (Google Safe Browsing API). The dataset covers 2,672 domains categorized as fake, unreliable, mixed, conspiracy, or satire, each assigned a composite credibility score between 0.0 and 1.0. CRED-1 is designed for on-device deployment in privacy-preserving browser extensions to enable client-side pre-bunking of misinformation at the content delivery stage. The entire pipeline is implemented in Python using only standard library modules and is fully reproducible from publicly available sources. The dataset and pipeline code are released under CC~BY~4.0 and archived on Zenodo.

cs.IR↗

arXiv2026

Eroding the Truth-Default: A Causal Analysis of Human Susceptibility to Foundation Model Hallucinations and Disinformation in the Wild

As foundation models (FMs) approach human-level fluency, distinguishing synthetic from organic content has become a key challenge for Trustworthy Web Intelligence. This paper presents JudgeGPT and RogueGPT, a dual-axis framework that decouples "authenticity" from "attribution" to investigate the mechanisms of human susceptibility. Analyzing 918 evaluations across five FMs (including GPT-4 and Llama-2), we employ Structural Causal Models (SCMs) as a principal framework for formulating testable causal hypotheses about detection accuracy. Contrary to partisan narratives, we find that political orientation shows a negligible association with detection performance ($r=-0.10$). Instead, "fake news familiarity" emerges as a candidate mediator ($r=0.35$), suggesting that exposure may function as adversarial training for human discriminators. We identify a "fluency trap" where GPT-4 outputs (HumanMachineScore: 0.20) bypass Source Monitoring mechanisms, rendering them indistinguishable from human text. These findings suggest that "pre-bunking" interventions should target cognitive source monitoring rather than demographic segmentation to ensure trustworthy information ecosystems.

cs.CY↗

arXiv2026

Industrialized Deception: The Collateral Effects of LLM-Generated Misinformation on Digital Ecosystems

Generative AI and misinformation research has evolved since our 2024 survey. This paper presents an updated perspective, transitioning from literature review to practical countermeasures. We report on changes in the threat landscape, including improved AI-generated content through Large Language Models (LLMs) and multimodal systems. Central to this work are our practical contributions: JudgeGPT, a platform for evaluating human perception of AI-generated news, and RogueGPT, a controlled stimulus generation engine for research. Together, these tools form an experimental pipeline for studying how humans perceive and detect AI-generated misinformation. Our findings show that detection capabilities have improved, but the competition between generation and detection continues. We discuss mitigation strategies including LLM-based detection, inoculation approaches, and the dual-use nature of generative AI. This work contributes to research addressing the adverse impacts of AI on information quality.

cs.CY↗

arXiv2026

Origin Lens: A Privacy-First Mobile Framework for Cryptographic Image Provenance and AI Detection

The proliferation of generative AI poses challenges for information integrity assurance, requiring systems that connect model governance with end-user verification. We present Origin Lens, a privacy-first mobile framework that targets visual disinformation through a layered verification architecture. Unlike server-side detection systems, Origin Lens performs cryptographic image provenance verification and AI detection locally on the device via a Rust/Flutter hybrid architecture. Our system integrates multiple signals - including cryptographic provenance, generative model fingerprints, and optional retrieval-augmented verification - to provide users with graded confidence indicators at the point of consumption. We discuss the framework's alignment with regulatory requirements (EU AI Act, DSA) and its role in verification infrastructure that complements platform-level mechanisms.

cs.CR↗

arXiv2026

The Verification Crisis: Expert Perceptions of GenAI Disinformation and the Case for Reproducible Provenance

The growth of Generative Artificial Intelligence (GenAI) has shifted disinformation production from manual fabrication to automated, large-scale manipulation. This article presents findings from the first wave of a longitudinal expert perception survey (N=21) involving AI researchers, policymakers, and disinformation specialists. It examines the perceived severity of multimodal threats -- text, image, audio, and video -- and evaluates current mitigation strategies. Results indicate that while deepfake video presents immediate "shock" value, large-scale text generation poses a systemic risk of "epistemic fragmentation" and "synthetic consensus," particularly in the political domain. The survey reveals skepticism about technical detection tools, with experts favoring provenance standards and regulatory frameworks despite implementation barriers. GenAI disinformation research requires reproducible methods. The current challenge is measurement: without standardized benchmarks and reproducibility checklists, tracking or countering synthetic media remains difficult. We propose treating information integrity as an infrastructure with rigor in data provenance and methodological reproducibility.

cs.CY↗

arXiv2024

Blessing or curse? A survey on the Impact of Generative AI on Fake News

Fake news significantly influence our society. They impact consumers, voters, and many other societal groups. While Fake News exist for a centuries, Generative AI brings fake news on a new level. It is now possible to automate the creation of masses of high-quality individually targeted Fake News. On the other end, Generative AI can also help detecting Fake News. Both fields are young but developing fast. This survey provides a comprehensive examination of the research and practical use of Generative AI for Fake News detection and creation in 2024. Following the Structured Literature Survey approach, the paper synthesizes current results in the following topic clusters 1) enabling technologies, 2) creation of Fake News, 3) case study social media as most relevant distribution channel, 4) detection of Fake News, and 5) deepfakes as upcoming technology. The article also identifies current challenges and open issues.

cs.CL↗

arXiv2024

Collaborative Cybersecurity Using Blockchain: A Survey

Collaborative cybersecurity relies on organizations sharing information to boost security, but trust management is a key concern. Decentralized solutions like distributed ledgers, particularly blockchain, are crucial for eliminating single points of failure. However, the existing literature on blockchain-based collaborative cybersecurity is limited, lacking comprehensive insights. This paper addresses this gap by surveying blockchain's role in collaborative cybersecurity from 2016 to 2023. It explores various applications, trends, and the evolution of blockchain technology, focusing on access control, data validation policies, underlying tech, and consensus mechanisms. A key finding is the fragmentation of the field with no dominant research group or venue. Many recent projects poorly select consensus protocols for their blockchain. To aid researchers and practitioners, this paper offers guidelines for choosing the right blockchain for specific purposes and highlights open research areas and lessons learned from past blockchain applications in collaborative cybersecurity, encouraging further exploration in this field.

cs.CR↗

arXiv2023

Real-Time Performance of OPC UA

OPC UA is an industry-standard machine-to-machine communication protocol in the Industrial Internet of Things. It relies on time-sensitive networking to meet the real-time requirements of various applications. Time-sensitive networking is implemented through various queueing disciplines (qdiscs), including Time Aware Priority, Multiqueue Priority, Earliest TxTime First, and Credit-Based Shaper. Despite their significance, prior studies on these qdiscs have been limited to a few. They have often been confined to point-to-point network topologies using proprietary software or specialized hardware. This study builds upon existing research by evaluating all these qdiscs in point-to-point and bridged topologies using open-source software on commercial off-the-shelf hardware. We first identify the optimal configuration for each qdisc and then compare their jitter, latency, and reliability through experiments. Our results show that open-source OPC UA on commercial off-the-shelf hardware can effectively meet the stringent real-time requirements of many industrial applications and provide a foundation for future research and practical deployments.

cs.NI↗

arXiv2020

A Generalized TDoA/ToA Model for ToF Positioning

Many applications require positioning. Time of Flight (ToF) methods calculate distances by measuring the propagation time of signals. We present a novel ToF localization method. Our new approach works infrastructure-less, without pre-defined roles like Anchors or Tags. It generalizes existing synchronization-less Time Difference of Arrival (TDoA) and Time of Arrival (ToA) algorithms. We show how known algorithms can be derived from our new method. A major advantage of our approach is that it provides a comparable or better clock error robustness, i.e. the typical errors of crystal oscillators have negligible impact for TDoA and ToA measurements. We show that our channel usage is for most cases superior compared to the state-of-the art.

cs.NI↗

arXiv2020

Clock Error Analysis of Common Time of Flight based Positioning Methods

Today, many applications such as production or rescue settings rely on highly accurate entity positioning. Advanced Time of Flight (ToF) based positioning methods provide highaccuracy localization of entities. A key challenge for ToF based positioning is to synchronize the clocks between the participating entities. This paper summarizes and analyzes ToA and TDoA methods with respect to clock error robustness. The focus is on synchronization-less methods, i.e. methods which reduce the infrastructure requirement significantly. We introduce a unified notation to survey and compare the relevant work from literature. Then we apply a clock error model and compute worst case location-accuracy errors. Our analysis reveals a superior error robustness against clock errors for so called Double-Pulse methods when applied to radio based ToF positioning

cs.NI↗

arXiv2019

Rule-Based Translation of Application-Level QoS Constraints into SDN Configurations for the IoT

In this paper, we propose an approach for the automated translation of application-level requirements regarding the logical workflow and its QoS into a configuration of the underlying network substrate. Our goal is to facilitate the integration of QoS constraints in the development of industrial IoT applications to make them more reliable. We follow an approach based on two semantic models: The first model allows to design the workflow of an IoT application and to express application-level QoS requirements on its interactions. The second model captures the configuration of a network and can be used as input to a north-bound interface of an SDN controller. Finally, we make use of rule-based semantic reasoning to automatically translate from the application requirements into SDN parameters.

cs.NI↗

arXiv2019

Deep Learning-aided Application Scheduler for Vehicular Safety Communication

802.11p based V2X communication uses stochastic medium access control, which cannot prevent broadcast packet collision, in particular during high channel load. Wireless congestion control has been designed to keep the channel load at an optimal point. However, vehicles' lack of precise and granular knowledge about true channel activity, in time and space, makes it impossible to fully avoid packet collisions. In this paper, we propose a machine learning approach using deep neural network for learning the vehicles' transmit patterns, and as such predicting future channel activity in space and time. We evaluate the performance of our proposal via simulation considering multiple safety-related V2X services involving heterogeneous transmit patterns. Our results show that predicting channel activity, and transmitting accordingly, reduces collisions and significantly improves communication performance.

cs.NI↗