Видавничий дім «Інтернаука» — на головну

Інше English Відкритий доступ

Explainability and Transparency in Artificial Intelligence: Methods, User Trust, and the Differences between Machine and Human Text

Mozolevskyi Dmytro

Міжнародний науковий журнал «Інтернаука», № 18, листопад 2023

Анотація

In recent years, artificial intelligence (AI) has reached an unprecedented level of development in the field of text generation, which has given rise to a set of important questions related to explainability, transparency, and user trust. This article provides a comprehensive analysis of modern methods for improving the comprehensibility of AI-generated texts, including linguistic, structural, and visual approaches. Particular attention is paid to factors influencing the formation of user trust in AI assistants, such as information accuracy, algorithm transparency, data protection, interface usability, and the degree of personalization. In addition, the paper thoroughly examines the ability of readers to differentiate machine-generated text from human text, considers characteristic markers of AI texts, and the effectiveness of modern detectors. The article consists of eight substantive sections, each of which represents an in-depth study of a specific aspect of the problem, supported by relevant data and three specialized graphs visualizing key trends and patterns.

Ключові слова: Explainable AI; AI transparency; large language models; LLM interpretability; AI-generated text detection; human-AI collaboration; algorithmic accountability; natural language generation; prompt engineering; bias mitigation; user trust; AI ethics; human perception; AI literacy; human-computer interaction

Як цитувати

Mozolevskyi D. Explainability and Transparency in Artificial Intelligence: Methods, User Trust, and the Differences between Machine and Human Text // International scientific journal "Internauka". — 2023. — №18. https://doi.org/10.25313/2520-2057-2023-18-11143

@article{internauka11143,
  author   = {Mozolevskyi, Dmytro},
  title    = {Explainability and Transparency in Artificial Intelligence: Methods, User Trust, and the Differences between Machine and Human Text},
  journal  = {Міжнародний науковий журнал «Інтернаука»},
  year     = {2023},
  number   = {18},
  doi      = {10.25313/2520-2057-2023-18-11143},
  url      = {https://www.inter-nauka.com/issues/2023/18/11143}
}
TY  - JOUR
AU  - Mozolevskyi, Dmytro
TI  - Explainability and Transparency in Artificial Intelligence: Methods, User Trust, and the Differences between Machine and Human Text
JO  - Міжнародний науковий журнал «Інтернаука»
PY  - 2023
IS  - 18
SN  - 2520-2057
DO  - 10.25313/2520-2057-2023-18-11143
UR  - https://www.inter-nauka.com/issues/2023/18/11143
LA  - en
ER  - 

Текст статті

Або відкрийте файл PDF

PDF, 863 КБ — Завантажити PDF

Як цитувати

Mozolevskyi D. Explainability and Transparency in Artificial Intelligence: Methods, User Trust, and the Differences between Machine and Human Text // International scientific journal "Internauka". — 2023. — №18. https://doi.org/10.25313/2520-2057-2023-18-11143

@article{internauka11143,
  author   = {Mozolevskyi, Dmytro},
  title    = {Explainability and Transparency in Artificial Intelligence: Methods, User Trust, and the Differences between Machine and Human Text},
  journal  = {Міжнародний науковий журнал «Інтернаука»},
  year     = {2023},
  number   = {18},
  doi      = {10.25313/2520-2057-2023-18-11143},
  url      = {https://www.inter-nauka.com/issues/2023/18/11143}
}
TY  - JOUR
AU  - Mozolevskyi, Dmytro
TI  - Explainability and Transparency in Artificial Intelligence: Methods, User Trust, and the Differences between Machine and Human Text
JO  - Міжнародний науковий журнал «Інтернаука»
PY  - 2023
IS  - 18
SN  - 2520-2057
DO  - 10.25313/2520-2057-2023-18-11143
UR  - https://www.inter-nauka.com/issues/2023/18/11143
LA  - en
ER  -