AI Scholarly Study
Abstract
AI has significantly affected communication fields, particularly text generation.
Differentiating between human and AI-generated content is critical for authorship in education, journalism, and content development.
Research focuses on linguistic tools to distinguish these texts in EFL proficiency assessment.
Research Methodology
Mixed approach combining qualitative and quantitative analysis.
Qualitative identifies linguistic markers; quantitative focuses on sentence length, vocabulary diversity, frequency, and syntactic complexity.
Key Findings
Distinctive features of AI texts: repetition, inappropriate style, inconsistent tone, lack of coherence, and low-quality referencing.
Identifies gaps in human expression in AI text.
Further research needed on AI's evolving linguistic markers in different domains.
Introduction
Importance of distinguishing human from AI-generated content has increased with advancements in AI tools.
Both types of content share similarities in conveying information and grammatical structure.
Indicators of AI-generated content include low-quality sources and unnatural language structures.
Linguistic Characteristics of Human-Created Discourse
Human texts exhibit grammatical understanding and may contain errors related to non-native usage.
Cohesion in human writing relies on cognitive processes influenced by experience and linguistic competence.
Non-native texts may show signs of unnatural language use affecting communication clarity.
Comparison and Analysis
A mixed methods research design identifies linguistic markers in both human and AI-generated texts.
Qualitative approach analyzes vocabulary, grammar, syntax, and style, while quantitative assesses diversity and complexity.
Emphasizes the need for integrity in referencing and citing in academic writing.
Limitations
Potential limitations include representativeness of texts and the evolving capabilities of AI.
Further studies should address and validate differences in linguistic markers.
Conclusions
Clear distinctions between human and AI-generated texts are necessary for maintaining academic integrity and fostering creativity.
Important to develop detection tools as AI models continue to advance.
Keywords
Discourse, Artificial Intelligence, Language Patterns, Style, Vocabulary, Grammar, Syntax, Structure, Cohesion.