How Are AI Pros Weighing ChatGPT’s Role in the World of Science?

New evolving and fast-developing AIs have started rising to fame in recent years, and with the recent invention of ChatGPT – an AI chatbot launched in late 2022 which gained over 100 million users within the first two months of launching – people from all around the world are starting to question the role of AI technology in their daily lives. Among these uncertainties is ChatGPT’s role in the world of science.

Will there be a huge impact of ChatGPT and developing AI in science? How will Chat GPT and developing AI contribute to modern science?

Biologists like Casey Greene and Milton Pividori see the tool as incredibly useful. After running their own research of using the tool for improving the quality and readability of academic manuscripts, Pividori concluded that ChatGPT, alongside other AI tools, is adding value to their lives by making helping scientists be “more productive as researchers.”

Meanwhile, other scientists are saying that they now regularly use LLMs (large language models like ChatGPT) not only to edit manuscripts but also to help them write or check code and to brainstorm ideas. As Nature.com reports, scientists like Almira Osmanovic Thunström, a neurobiologist at Sahlgrenska University Hospital in Gothenburg, says the use of ChatGPT will keep scientists from needing to “sit and write long introductions for grant applications.”

That being said, many scientists and researchers have pointed out a few disadvantages of AI tools. These are in the form of inaccurate information and fictitious citations presented by AI machines, giving scientists a reason to rely less on AI and more on their own manual research and cross-referencing. Can researchers confidently pinpoint ChatGPT’s role in the world of science?

Ganesh Gopalan, CEO and co-founder of Gnani.ai, a conversational AI platform with a focus on customer experience enhancement, weighs in on the impact of ChatGPT and other developing AI tools in the world of science and the difference it can make in research fields.

Ganesh Gopalan’s Thoughts

“ChatGPT is going to help researchers automate all their routine activities such as writing publications, reviewing documents are automatically creating extracts.

These articles typically warn of the pitfalls in terms of biases that ChatGPT could throw up and the possible inaccuracies. So, what I thought the best thing to evaluate this is to ask ChatGPT the same question. When I ask ChatGPT the same questions, I got a lot more results such as the influence on drug discovery, the influence on analyzing large pieces of scientific data and finding the relationships.

So essentially, I guess ChatGPT has a lot of benefits for science but you’ve gotta be a little careful on the biases and also on the accuracy of the data.”

Article written by Azam Saghir.