What is Wrong with Generative AI? How to Communicate with Humans in Sentence Problems: A Case Study using Gemini and a Conversational Approach
generativeai presents challenges to the scientific community But it can also enhance the quality of our work. These tools can bolster our capabilities in writing, reviewing and editing. They preserve the essence of scientific inquiry — curiosity, critical thinking and innovation — while improving how we communicate our research.
Context is king. You can’t expect generative AI — or anything or anyone, for that matter — to provide a meaningful response to a question without it. When using a chatbot to help you with your paper, begin by outlining the context. What is your main argument in your paper? Bullet points can work in any format you choose. The generative artificial intelligence of your choice will find this information useful. I typically use ChatGPT, made by OpenAI in San Francisco, California, but for tasks that demand a deep understanding of language nuances, such as analysing search queries or text, I find Gemini, developed by researchers at Google, to be particularly effective. The large language models made by Mixtral are ideal when you can’t always be reached by a human but still need help.
The first reply may not be perfect, but it should be remembered that the process is iterative and collaborative. You might need to refine your instructions or add more information, much as you would when discussing a concept with a colleague. It’s the interaction that improves the results. Don’t hesitate to say, “This isn’t what I meant” when something doesn’t hit the mark. Let’s adjust this part.” It is clearer, but we need to make a better transition to the next section.
After evaluating a paper, I can get it to draft a letter on the basis of these notes. Highlight the manuscript’s key issues and clearly explain why the manuscript, despite its interesting topic, might not provide a substantial enough advancement to merit publication. Don’t use jargon. Be direct. Maintain a professional and respectful tone throughout.” It may take a few attempts to get the tone and content just right.
I’ve found that this approach both enhances the quality of my feedback and helps to guarantee that I convey my thoughts supportively. The result is a more positive discussion between the two parties.
How accurate can Turnitin be? Detecting Artificial Intelligence in the Writing of Anglo-Bahamian Students using an Empirical Detector
Teachers want to hold students accountable for using artificial intelligence without permission. But that requires a reliable way to prove AI was used in a given assignment. Instructors have tried to find their own solutions to detecting artificial Intelligence in writing, using messy, untested methods to enforce rules, and distressing students. Some teachers are making their grades by using generative artificial intelligence.
The risk of bias is inherent in detection tools. A study published in the American Journal of Sociology found a false positive rate of 61.3 percent when evaluating the test of English as a foreign language. The study did not look at the author’s version. The company says that it has training on detecting the writing of English language learners and native English speakers. A study published in October found that Turnitin was among the most accurate of 16 AI language detectors in a test that had the tool examine undergraduate papers and AI-generated papers.
