Professional journalism standards in news processing using AI-based large language models compared to human processing:

This exploratory study aims to assess the extent of compliance with professional journalism standards (Topicality, Reliability, Balance of opinion, separation of facts and opinions, accuracy of information, and completeness of information) in the responses of large language models represented in Chat GPT and compared to the treatment of humans (journalists) and the extent to which the content is supported by arguments, proofs, evidence, figures and statistics, and this study relied on the experimental method, and the tools were the content analysis and the systematic comparison tool, to compare serious and light news in the New York Times. This study reached a set of results, the most important of which are: The superiority of news treatment of humans over the responses of the large language model Chat GPT in adherence to professional journalism standards, and the biggest feature was the so-called "hallucinations", so the information received seemed logical, but it was misleading and fabricated. The large chat GPT language model did not adhere to the journalistic form recognized in formulating news from the introduction, body and conclusion, and there were no clear differences between serious and light news in terms of adherence to journalism standards, the large language model did not base the information contained in its responses to sources, don’t use any quotes. Large language models need more training to accurately mimic human writing and adhere to professional journalism standards.