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.