Carding 4 Carders
Professional
AI and science: Two best friends or competitors?
Radiologist Domenico Mastrodicasa from the University of Washington School of Medicine in Seattle uses ChatGPT as a "sounding board" when writing scientific articles, which allows him to quickly create ready-to-publish manuscripts. Mastrodicasa is one of many researchers experimenting with generative AI tools for writing text or code. He finds ChatGPT particularly useful for suggesting clearer ways to express his ideas.
Many scientists expect that generative AI tools will become permanent assistants when writing manuscripts, expert reports, and grant applications. Scientific publishers are already experimenting with generative AI in scientific search tools and for editing and quickly summarizing articles.
Michael Eisen, a computational biologist at the University of California, Berkeley, believes that generative AI tools can fundamentally change the nature of a scientific paper. However, there are concerns about inaccuracies and false information. Publishers are concerned that the increased use of such tools may lead to a large number of low-quality or erroneous manuscripts.
Laura Fifam of IOP Publishing expressed concerns about the potential consequences of generative AI. Daniel Hook, CEO of Digital Science, emphasized that publishers have every reason to worry about the possibility of creating low-quality articles using AI.
Publishers address the issue either by completely banning the use of LLM (as the Science publisher did), or, in most cases, by insisting on transparency. Some editors worry that generative AI can be used to create fake but compelling articles.
However, many researchers believe that generative AI tools can bring great benefits to science, especially for researchers for whom English is not their native language. Such tools can help improve equality in science.
Generative AI tools have the potential to change the way research is published and distributed. Patrick Maino, a senior scientist in machine learning at the AI Institute of Quebec, suggests that research will be published in a way that makes it easy for machines to read, rather than humans.
Radiologist Domenico Mastrodicasa from the University of Washington School of Medicine in Seattle uses ChatGPT as a "sounding board" when writing scientific articles, which allows him to quickly create ready-to-publish manuscripts. Mastrodicasa is one of many researchers experimenting with generative AI tools for writing text or code. He finds ChatGPT particularly useful for suggesting clearer ways to express his ideas.
Many scientists expect that generative AI tools will become permanent assistants when writing manuscripts, expert reports, and grant applications. Scientific publishers are already experimenting with generative AI in scientific search tools and for editing and quickly summarizing articles.
Michael Eisen, a computational biologist at the University of California, Berkeley, believes that generative AI tools can fundamentally change the nature of a scientific paper. However, there are concerns about inaccuracies and false information. Publishers are concerned that the increased use of such tools may lead to a large number of low-quality or erroneous manuscripts.
Laura Fifam of IOP Publishing expressed concerns about the potential consequences of generative AI. Daniel Hook, CEO of Digital Science, emphasized that publishers have every reason to worry about the possibility of creating low-quality articles using AI.
Publishers address the issue either by completely banning the use of LLM (as the Science publisher did), or, in most cases, by insisting on transparency. Some editors worry that generative AI can be used to create fake but compelling articles.
However, many researchers believe that generative AI tools can bring great benefits to science, especially for researchers for whom English is not their native language. Such tools can help improve equality in science.
Generative AI tools have the potential to change the way research is published and distributed. Patrick Maino, a senior scientist in machine learning at the AI Institute of Quebec, suggests that research will be published in a way that makes it easy for machines to read, rather than humans.
