Mykyta Mudryi | Searcharxiv

arXiv2026

Entropy of Ukrainian

In natural language processing, the entropy of a language is a measure of its unpredictability and complexity. The first study on this subject was conducted by Claude Shannon in 1951. By having participants predict the next character in a sentence, he was able to approximate the entropy of the English language. Several follow-up studies by other authors have since been conducted for English, and one for Hebrew. However, to date, Shannon's experiment has never been conducted for Ukrainian. In this paper, we perform this experiment for Ukrainian by recruiting 184 volunteers using social media channels. We rely on techniques used for English to approximate the entropy value of Ukrainian. The final result is an upper bound of $H_{upper}\approx1.201$ bits per character. We compare this to the performance of current Large Language Models. The methods and code used are also documented and published, along with a discussion of the main challenges encountered.

cs.CL↗

arXiv2026

Global AI Governance Overview: Understanding Regulatory Requirements Across Global Jurisdictions

The rapid advancement of general-purpose AI models has increased concerns about copyright infringement in training data, yet current regulatory frameworks remain predominantly reactive rather than proactive. This paper examines the regulatory landscape of AI training data governance in major jurisdictions, including the EU, the United States, and the Asia-Pacific region. It also identifies critical gaps in enforcement mechanisms that threaten both creator rights and the sustainability of AI development. Through analysis of major cases we identified critical gaps in pre-training data filtering. Existing solutions such as transparency tools, perceptual hashing, and access control mechanisms address only specific aspects of the problem and cannot prevent initial copyright violations. We identify two fundamental challenges: pre-training license collection and content filtering, which faces the impossibility of comprehensive copyright management at scale, and verification mechanisms, which lack tools to confirm filtering prevented infringement. We propose a multilayered filtering pipeline that combines access control, content verification, machine learning classifiers, and continuous database cross-referencing to shift copyright protection from post-training detection to pre-training prevention. This approach offers a pathway toward protecting creator rights while enabling continued AI innovation.

cs.CY↗

arXiv2026

Copyright in AI Pre-Training Data Filtering: Regulatory Landscape and Mitigation Strategies

The rapid advancement of general-purpose AI models has increased concerns about copyright infringement in training data, yet current regulatory frameworks remain predominantly reactive rather than proactive. This paper examines the regulatory landscape of AI training data governance in major jurisdictions, including the EU, the United States, and the Asia-Pacific region. It also identifies critical gaps in enforcement mechanisms that threaten both creator rights and the sustainability of AI development. Through analysis of major cases we identified critical gaps in pre-training data filtering. Existing solutions such as transparency tools, perceptual hashing, and access control mechanisms address only specific aspects of the problem and cannot prevent initial copyright violations. We identify two fundamental challenges: pre-training license collection and content filtering, which faces the impossibility of comprehensive copyright management at scale, and verification mechanisms, which lack tools to confirm filtering prevented infringement. We propose a multilayered filtering pipeline that combines access control, content verification, machine learning classifiers, and continuous database cross-referencing to shift copyright protection from post-training detection to pre-training prevention. This approach offers a pathway toward protecting creator rights while enabling continued AI innovation.

cs.CY↗

arXiv2025

The Hidden Dangers of Browsing AI Agents

Autonomous browsing agents powered by large language models (LLMs) are increasingly used to automate web-based tasks. However, their reliance on dynamic content, tool execution, and user-provided data exposes them to a broad attack surface. This paper presents a comprehensive security evaluation of such agents, focusing on systemic vulnerabilities across multiple architectural layers. Our work outlines the first end-to-end threat model for browsing agents and provides actionable guidance for securing their deployment in real-world environments. To address discovered threats, we propose a defense in depth strategy incorporating input sanitization, planner executor isolation, formal analyzers, and session safeguards. These measures protect against both initial access and post exploitation attack vectors. Through a white box analysis of a popular open source project, Browser Use, we demonstrate how untrusted web content can hijack agent behavior and lead to critical security breaches. Our findings include prompt injection, domain validation bypass, and credential exfiltration, evidenced by a disclosed CVE and a working proof of concept exploit.

cs.CR↗