Every researcher has to write himself because the programs don't know what you're thinking

@Ramtin Zargari Marandi; @Bakhodir S. Okhundedaev thank you for your responses.

ChatGPT would not be able to provide humans with credible research work, however, there could be the possibility of AI to direct human researchers to relevant information needed to direct the search for relevant literature, for example.

I think that the use of ChatGPT in academic writing poses a significant challenge to researchers. ChatGPT can generate reasonable abstracts which may be better than those written by researchers in some cases. This raises questions about the acceptability and credibility of such artificial intelligence generated works in the academic community. As researchers, we should be concerned about this development, as it may lead to a devaluation of our work. It is also important to consider how ChatGPT may be used as a tool to assist in research rather than as a replacement for it. The future of researchers may involve a more collaborative approach to research, involving the use of both human and artificial intelligence. Researchers must adapt to this new environment and develop skills to best utilize such technologies to their advantage. Overall, ChatGPT presents a challenge to researchers, but also an opportunity to work with new technologies and develop new skills. The acceptability and credibility of ChatGPT in academic writing may depend on the level of oversight, control, and quality assurance applied by the academic community.

Thank you for your contribution @Peter Donkor. Your contribution is very well appreciated.

Very recently, ChatGPT was praised across the world including among some academics. It is important to note that this would not be the first AI application that has been incorporated in academia. Some e-journals, libraries, and reference managers, just to mention a few have been with us for some decades now. The reference managers, for example, would not do a perfect job without the user's experience in academic writing.

Although many academics are already concerned about ChatGPT, time will tell how it is going to impact academia.

Great input from @Ramtin Zargari Marandi.

Thank you for your invaluable and insightful contribution. 👍🏼

The link below might be interesting and insightful:

https://www.nature.com/articles/d41586-023-00107-z?utm_source=Nature+Briefing&utm_campaign=2fd6fa2583-briefing-dy-20230120&utm_medium=email&utm_term=0_c9dfd39373-2fd6fa2583-45203210

Thank you for your time.

There will undoubtedly be some significant challenges.

ChatGPT is a powerful language model that can generate text that is similar in quality to that written by humans. This can be beneficial in a number of ways, such as helping researchers to quickly generate abstracts or other written materials. However, it also raises some concerns about the potential for the model to be used in ways that could undermine the credibility of research or academic writing. Researchers will likely need to adapt to this new technology and develop strategies for ensuring the quality and credibility of their work in the face of increasing automation. Ultimately, in my opinion the future of researchers will likely involve working in collaboration with AI tools like ChatGPT, rather than being replaced by them.

Great input here@Emiliana Minenna

Thanks for sharing your thoughts here🙏🏽.

ChatGPT is weak AI. Any experienced scientist would be able to pick up the shallow answers.

You are showcasing yesteryear's AI. No big deal here, just simple Deep Learning optimization techniques are used to refine the answers to best match a possible solution. The problem with any weak AI is they are incapable of understanding if the solution is best. Turing computation is incapable of deciding on quality and hence poor descriptor of creative papers.

See

https://www.researchgate.net/project/nonTuring-computation-in-the-brain-the-conscious-act-of-understanding-uncertainty

Eda Alemdar

The style looks good, but the content would require so much work that makes it not worth the effort. Some information is missing, even on basic concepts, like insects anatomy, and there is plenty of fake information. I tried to ask for a summary of the topic I'm writing about and it looked too good to be true, in fact a lot information was generated by chatgpt. I also noticed recurring patterns in the writing style and that it frequently uses double spaces.

I'm not sure if it can improve with interaction. The time that you might use training it could be instead used for doing actual research and writing.

It also cannot improve text so easily. I try feeding it paragraphs asking to rephrase better, but the meaning was most often lost or the it didn't sound better, at least to me.

I'm not sure it can help with directing the search for relevant literature either.

Overall, I think that it's still too weak to be used and, most importantly, academia and publishing are riddled with issues that we should fix before introducing such a tool.

A.I. Like ChatGPT Is Revealing the Insidious Disease at the Heart of Our Scientific Process

An AI program cannot be an author. A violation of these policies will constitute scientific misconduct no different from altered images or plagiarism of existing works...

These might seem like gentle warnings, but to academics who submit research papers to peer-reviewed journals like Science and Nature, the specter of being charged with research misconduct—potentially a career-wrecking accusation—for using A.I. is about as subtle as an air-raid siren. It’s a bright red neon sign saying, in all caps, “Don’t go there.”

https://slate.com/technology/2023/01/ai-chatgpt-scientific-literature-peer-review.html

Thank you All for your wonderful contribution to this discussion. Great thanks to Professor Ljubomir Jacić Nicolò M. Villa R. Poznansky Emiliana Minenna Mohammad Hamad Al-khresheh . All your contributions to this discussion are so much appreciated!

Ramtin Zargari Marandi Peter Donkor Mohammad Hamad Al-khresheh Emiliana Minenna R. Poznansky Nicolò M. Villa Ljubomir Jacić

Thank you so much and well done!

[Are we in trouble having learned that ChatGPT can write reasonable abstracts almost better than what many researchers can do? How challenging is this to serious-minded researchers? How concerned should we be? What does the future of researchers look like? What do you think about the acceptability and credibility of ChatGPT in academic writing?]

That was the discussion question and it was not in any way a claim or a form of Eureka something new is discovered. If we follow science news closely, we would see how ChatGPT is shaping headlines.

However, AI of this nature has been in use for ages without anyone making noise about it. How researchers conduct literature searches has been with the help of similar tools.

Detecting plagiarism, applying literature management software, and analysing texts, audio, and video data in qualitative research have all seen application software of a similar nature.

The post was aimed at generating interesting and intellectual discussion while also sharing our opinions and it was not aimed at displaying any concrete claim about a never-heard-of thing.

Very interesting contributions! Thank you All! Best wishes for the weekend!

For small tasks ChatGPT does a reasonable job but asking it to write a paper that I would take with a large grain of salt. Currently ChatGPT-3 has about 175 billion parameters, the next increment ChatGPT 4 is slated to have over 1 Trillion .. the performance is expected to improve but with so many parameters to tune it feels more like the mother-of-all lookup tables.

I am concerned that it may facilitate cheating. There will be a dark side to ChatJPT.

Definitely ... the problem will be to how to uncover this.

Very insightful Mohammad Hamad Al-khresheh and Nicolangelo Iannella

There we go now. I think we might want to prove something to those who would have dismissed the fact that ChatGPT can actually produce something closer to what is difficult to distinguish from that made by humans. I think, as Nicolangelo Iannella rightly stated, the problem of how to uncover this would be the next challenge. However, no serious-minded researcher would take the work produced by ChatGPT, I guess.

Thanks so much, Mohammad Hamad Al-khresheh and Nicolangelo Iannella for your contributions.

The artificial intelligence (AI) chatbot ChatGPT shows how science “now faces a reckoning induced by AI technology infringing on its most dearly held values, practices and standards”, argue five researchers who specialize in psychology, cognition and AI. The authors delve into the complexities of accuracy and misinformation, how to bestow attribution and credit, and how AI might ultimately remake the scientific enterprise...

https://www.nature.com/articles/d41586-023-00288-7

Today research publishing has developed into an efficient digital process, which not only saves time in publishing papers but also helps it reach out to the audience in the scientific world. In this blog, we will discuss how AI in research has improved and enhanced the process of research publishing and how does the future of publishing look through an AI lens...

AI-driven tools are assisting researchers in improving their research content quality by providing assisted proofreading and language editing services. With the increased number of academic research paper publications, there has been a massive growth in research database and AI-tools have enabled scanning through millions of scholarly database records and instantly retrieve the required research paper.

AI tools are assisting researchers in overcome the language related barriers and helping researchers create content which is free of grammatical errors. With the innovative tools and user friendly applications, AI is the leading game-changer in the publishing industry...

https://www.enago.com/academy/research-publishing-advent-of-ai/

Science publishing is not about the haphazard collection of papers. It is bringing something NEW onto the dining table of research. No weak AI can do that. Why? Here is my appraisal from another thread.

Google has made claims that Lamda AI is sentient. They fail miserably for three reasons. First, DeepMind architecture will never produce the semantic information necessary for sentience. Second, the conversation between humans and Lamda AI was all in agreement. The AI did not offer a counterargument or show a rebellious vent. It shows no sentience!

Finally, consciousness is built from many different meanings attributed to the lability of the informational structure. The consciousness code is chaotic, and there are infinite possibilities for different meanings to arise. The Lamda AI based on DeepMind has a finite repertoire of possibilities, limiting it to errors in situations that arise outside of the expected.

The problem with Google AI is that it has invested 500 million in DeepMind and will exploit it until they finally see that sentience cannot be "squeezed out of a rock" using Deep learning.

Strong AI will emerge based on the theory of non-integrated information, and possibly new companies will replace Google.

Google should invest in people not DeepMind for fresh ideas!

"AI writing tools are developing at a rapid pace and so is Turnitin’s technology to detect these emerging forms of misconduct. Recently, we shared with you that we have technology that can detect AI-assisted writing and AI writing generated by tools such as ChatGPT. Today, we want to introduce you to our AI Innovation Lab to give you a first-hand glimpse of what our technology (in development) can do.

Turnitin has been working on AI-powered solutions for several years now, and now we’d like to take you along on the ride. Watch this short demo where David Adamson, an AI scientist at Turnitin and a former high school teacher, walks you through our AI writing detection capability..."

https://www.turnitin.com/blog/ai-writing-the-challenge-and-opportunity-in-front-of-education-now

https://www.turnitin.com/blog/meet-the-passionate-people-who-work-on-turnitin-ai

https://www.turnitin.com/blog/sneak-preview-of-turnitins-ai-writing-and-chatgpt-detection-capability

I recommend to read this manuscript.

ChatGPT in Academic Writing and Publishing: A Comprehensive Guide

Scientific writing is a difficult task that requires clarity, precision, and rigour. It also involves a large amount of research, analysis, and synthesis of information from various sources. However, scientific writing is also hard, time-consuming, and susceptible to errors. Advanced artificial intelligence (AI) models, such as ChatGPT, can simplify academic writing and publishing. ChatGPT has many applications and uses in academic and scientific writing and publishing such as hypothesis generation, literature review, safety recommendations, troubleshooting, tips, paraphrasing and summarising, editing, and proofreading, journal selection, journal style formatting, and other applications.

In this book chapter, we will discuss the main advantages, examples, and applications of ChatGPT in academic and scientific writing from research conception to publishing.

Read online.

https://achtago.com/publications/list/chatgpt-in-academic-writing-and-publishing/

Download full text

https://www.researchgate.net/publication/369817340_Chapter_2_ChatGPT_in_Academic_Writing_and_Publishing_A_Comprehensive_Guide

“The human-like quality of the text structure produced by ChatGPT can deceive readers into believing it is of human origin. It is now apparent, however, that the generated text might be fraught with errors, can be shallow and superficial, and can generate false journal references and inferences. More importantly, ChatGPT sometimes makes connections that are nonsensical and false.”

https://mediterraneanworld.wordpress.com/2023/03/02/teaching-and-chatgpt/

Dear respected researchers

We have worked on a paper on that particular issue (ChatGPT and Academic research)

Title: "ChatGPT and Academic Research: A Review and Recommendations Based on Practical Examples"

The article's link:

https://journals.cspc.edu.ph/index.php/jemds/article/view/175

I hope you will enjoy the article and it will be helpful for academic people.

Regards.

I never got convinced with ChatGPT's answers to common questions. It might provide too general answer or paraphrase a text but does not show a reliable text that you can just use as it is. Of course, we expect it to develop further, but I still believe that it can be a tool that assists in research but not a full replacement.

https://www.theguardian.com/technology/2023/may/02/geoffrey-hinton-godfather-of-ai-quits-google-warns-dangers-of-machine-learning

As the generative AI (GAI) technology evolves into providing more human-like responses in text and image generation, academic community members must understand the influence of GAI in educational and scholarly writing. Despite the benefits of time-saving, ease-of-use, and accessibility, the issues of inaccuracies, lack of originality, and poor logical flows and elaborations were being raised by critics. These concerns were brought up on top of the ethical questions on authorships, biases, discrimination, and plagiarism. This research tested a few AI text generators and AI detectors and presented the findings using the utilitarianism theoretical framework and SWOT approach... (PDF) "CHATBOTS IMPACT ON ACADEMIC WRITING". Available from: https://www.researchgate.net/publication/370561426_CHATBOTS_IMPACT_ON_ACADEMIC_WRITING [accessed May 19 2023].

Artificial Intelligence–Generated Research in the Literature: Is It Real or Is It Fraud?

AI can write scientific articles by scouring the literature online, collecting phrases and images from articles already published and developing a totally new article that is nothing more than an assembled puzzle of cut-and-pasted published content—AI is high-tech plagiarism; it is fraud!

Article Artificial Intelligence-Generated Research in the Literature...

We have seen what ChatGPT can do: write a birthday poem in the manner of Emily Dickinson, write a rap for Napoleon lamenting Waterloo, write a play about the Iroquois hunting dinosaurs on the moon. Everything is everywhere all at once in the AI era. Chatbots hallucinate, making up facts and legal cases and works of scholarship that don’t exist...

https://lareviewofbooks.org/article/poetry-biography-and-the-unknowable-on-two-new-books-about-phillis-wheatley/

Bias in generative AI models can have significant implications for scientific writing and research. Therefore, when these models are trained on biased datasets, the generated content may reflect and perpetuate those biases, thereby leading to skewed conclusions and misleading information...

https://www.enago.com/academy/generative-ai-ethics-in-academic-writing/

The write algorithm: promoting responsible artificial intelligence usage and accountability in academic writing

OpenAI’s ChatGPT, a leading example, excels at analysing text and generating content based on user input. These breakthroughs have profound implications for academic writing, attracting the attention of journals worldwide. While the pros and cons of adopting these technologies have been extensively debated, the responsible implementation and transparent documentation of their use remain relatively overlooked. This Editorial seeks to fill this gap...

The updated ICMJE recommendations thoughtfully address AI’s potential consequences in scholarly publishing. Adoption must proceed cautiously, considering limitations in generating misinformation. Transparency, accountability, and ethical use should guide the development and integration of AI-assisted technologies. While AI can assist in various processes, human creativity, curiosity, and ingenuity remain distinctive and invaluable qualities in science and scholarship that will serve as the bedrock of these disciplines for years to come...

Article The write algorithm: promoting responsible artificial intell...

More than 15% of the 1,600 researchers who took part in a Nature survey use AI tools such as ChatGPT to help them write grant proposals. The fact that AI can do much of this work makes a mockery of the process, says robotics researcher Juan Manuel Parrilla Gutierrez...

https://www.nature.com/articles/d41586-023-03238-5

Response from ChatGPT

The capabilities of AI models like ChatGPT to generate high-quality abstracts and even research papers can be both exciting and challenging for the research community. Here are some key points to consider:

It's important for the research community to maintain a critical and discerning approach to AI-generated content and to adapt to the changing landscape of research. AI can be a valuable tool to augment and expedite research, but it should be used responsibly and in conjunction with traditional research methods. The future of researchers will likely involve a synergy between human expertise and AI assistance to advance scientific progress.

Tristan Charles-Dominique You bring up some good points but if you are going to use ChatGPT to write your abstract then I would argue that it would be faster to write your own. Where I see ChatGPT being of use is to provide summaries of research you are interested in. Letting ChatGPT write your paper brings about the question of authorship. Apart from that, ChatGPT is having a large impact with high school students and preparing essay and reports, where many ask for summaries of pages they should read and then they expand and change the text provided by ChatGPT.

It is maybe good to read the following paper (73 authors):

Opinion Paper: “So what if ChatGPT wrote it?” Multidisciplinary perspectives on opportunities, challenges and implications of generative conversational AI for research, practice and policy

I do also question the authorship of ChatGPT as Nicolangelo Iannella

Article “So what if ChatGPT wrote it?” Multidisciplinary perspective...

There is software to detect ChaptGPT authored works offered to publishers.

My guess is chaptGPT offers blend type papers that are abundant. A human who has no passion for writing a paper will write a blend paper but not abundant. The abundance gives the paper more worth. This is a human flaw that big must be good. All that is required is good editorial scrutiny or gut feeling.

all the best,

R

Nicolangelo Iannella

What is the name of this ChatGPT detector?

The authors did not named yet this ChatGPT detector. you may find their article about this detector.

Article Accurately detecting AI text when ChatGPT is told to write l...

Full text: https://pdf.sciencedirectassets.com/776855/AIP/1-s2.0-S2666386423005015/main.pdf?X-Amz-Security-Token=IQoJb3JpZ2luX2VjEJz%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEaCXVzLWVhc3QtMSJHMEUCIQDlGAv%2Fg34GaLPcFbRsIZw34ex5a4ppKhYaQapnOhn%2BDgIgdAXrM9FT%2BNvxHKfAOf%2B8rIumJ%2Bm0V1WzzPquGkmcCSMqvAUI1f%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FARAFGgwwNTkwMDM1NDY4NjUiDMj9uErztpAAvdpXpiqQBYk4SbV1pxUFjkF6Br1okukA4Hzwt1EHceg1zUmsbg0bb%2Fvco5yp%2BoLFBV8zs4aa9cZnpBBPrxSD0GaAtD22odLlcdtQccvphBk69IYFupZF8TzLP81Pu7tmL9x1Lcj24zBMU0OAHw4HWrTlRPDW4msaUl31kr7SiDDL4Q%2BOBV3xu4cY8iIHctbpMembOl6lPYn3%2FfBQzb3%2FiYtZdrQBUhaUG8nKruINeH6b4jE6Lq3%2BgRGhwZ5AtXi9HlT%2BGgKbx3%2FrvKJoxPlxVyxwqSI3BcrBvmZPUUkpFegx5LezSKEh%2FjP3czKqsEzd7FSEnr9yHVBgD6%2F3%2F8yh%2BIIEiL3k7IF5%2BdCvV6XxJggkw7LY31rTEy3dMj5fRsLRqyw0UyclVLixmj%2FbMzUxUykOjKI%2F3GOcLbX13UQJKntKMQJM%2BPtLe1xLoPZetfLkQIHW34UizBGkydSe6QGM0Fx8584uStScT0In0UYc4xuCRpBGPv%2FhGUR75aqekIQdnduR5r4ZN7GFQeeluNCcnF8CjapzJGaXbolYrXm1s%2BejMBDlVWIchdX4V3MMKwxRn4BZYd9C8DoJ44BRbimMUSrjBpkFEtpUYS7YbcDTWzQ3wM8LPk9tghFCTg1gn7V3Z1t04GRvJ4mSmHzQvGpo7kuKKgNONWCjx55b%2F0eLtGIxGlHjAi0imwQyGwsztVpgceM%2FYrp4rAAjbBkXisdW55Rgaau4DMsHGocXxQYhRpWJuRT8NDp9TDFGcdXhwZIySWNRDAJ%2FfXUZY5YwFpAPWRYR%2B5QWrYNnn5S11O9T1eIvdT89Exeqa3ZOCCIh9lwSwz5QwNo5RgBHEpDmiuZqY2EP54utc0euJTtsnxEE23nrr6K6rBX0MP2Os6oGOrEBOMuZRnW2LR6RaOoF7JEKrjsJX2SDTWLlqJXUTWmljGA9bbAhAfcc6ao8ACldOsP%2BtsSov3R38UkKbxxqzwzPQCUD0kMhfkPoOSiCc62ewZhLMhLaEzJ0n%2B%2F3EliJLFYFvkHrlhp%2FA2Qf9wAgKCZBM8l9DDrczK%2BbilEQzt1aNfmMhd8r8BaQ1VgWXqkd9nr4cjI%2FToMAa%2F6RymOkujObL%2FNj9Fez7HSjzLWKXyAqq4Qy&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20231109T131230Z&X-Amz-SignedHeaders=host&X-Amz-Expires=299&X-Amz-Credential=ASIAQ3PHCVTYSES556JB%2F20231109%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Signature=beefd75dd9c200b6eb59eec7f49fdc6e5d9d8b48f4d2cc2254ac5aa2b1865aa8&hash=f62e8d553153f22c01d656a910bfa43994f926754cf0eda151200444dd70d676&host=68042c943591013ac2b2430a89b270f6af2c76d8dfd086a07176afe7c76c2c61&pii=S2666386423005015&tid=spdf-7486762a-69ed-4cf7-8ae3-d66cc66e8ea3&sid=cc2c005b3864674116684d8073371097e5b1gxrqb&type=client&tsoh=d3d3LnNjaWVuY2VkaXJlY3QuY29t&ua=08155c575553530507575f&rr=82364ec71b63c25e&cc=rs

AI writes summaries of preprints in bioRxiv trial

Large language model creates synopses of papers aimed at various reading levels to help scientists sift through the literature...

https://www.nature.com/articles/d41586-023-03545-x

Despite the issues associated with AI-generated content, I believe that AI has the potential to enhance the text readability for non-English speakers. Rather than solely relying on AI for text generation, we can seek its assistance to improve readability. This approach could democratize scientific publication for the Third World. There is a lot of discussion with this issue for near future.

The use of artificial intelligence to improve the scientific writing of non-native english speakers

"Scientific writing in English is a daunting task for non-native English speakers. The challenges of writing in a foreign language are evident in the scientific literature where texts by non-native English-speaking scientists tend to be less clear and succinct, contain grammatical errors, and are often rejected by prestigious journals...

We conducted a non-systematic review of the most recent literature using the terms “Artificial Intelligence,” “Scientific Writing,” and “Non-English Speaking” to create a narrative review...

Artificial intelligence can be a solution to improve scientific writing, especially for non-native English-speaking scientists. Artificial intelligence can assist in the search for pertinent scientific papers, generate summaries, and help with the writing of different sections of the manuscript, including the abstract, introduction, methods, results, and discussion. Artificial intelligence-based programs can correct grammatical errors and improve writing style, both of which are particularly helpful for non-native English speakers...

Article The use of artificial intelligence to improve the scientific...

In line with the current discussion I would like to contribute with a note on sustainability: https://m.youtube.com/watch?v=eXdVDhOGqoE&t=525s&pp=ygUaU3VzdGFpbmFiaWxpdHkgYW5kIGNoYXRndHA%3D, watch the first part. This is a crucial aspect yet purposely neglected!

ChatGPT detector’ catches AI-generated papers with unprecedented accuracy," using "features of writing style to distinguish between human and AI authors...

https://www.nature.com/articles/d41586-023-03479-4