Depositphotos
The Massachusetts Institute of Technology (MIT) in the United States has retracted a high-profile article published last year about advantages of using artificial intelligence in scientific papers.
MIT is now demanding that the article be removed from public access and no longer published. This is a study entitled «Artificial Intelligence, Scientific Discovery and Innovation».
Last year, this article attracted wide attention from both scientists and the media. Its author Aidan Toner-Rogers based on the results of his research, claimed that scientists, who used AI in their work, were more productive than those who did not.
The paper also claimed that researchers who made more discoveries using AI, however, they were not satisfied with the results of their work. Doubts about the results in the scientific article «Artificial Intelligence, Scientific Discoveries and Innovations» were first raised by a scientist and programmer with a degree in materials science. He addressed MIT professors with questions about the mechanism of AI and an accurate assessment of how the model used has spurred innovation.
His appeal was forwarded to the MIT management. After checking the scientific material, the institute stated that they were no longer sure of the reliability of the material’s origin. However, MIT refused to disclose what exactly the problem was, citing student privacy laws and its own internal policies.
However, Aidan Toner-Rogers, the author of the article, no longer works at the university, and MIT demanded that the article be removed from the arXiv preprint site. The institute also postponed the material from consideration in the Quarterly Journal of Economics, where it was sent for evaluation and possible publication.
The researchers concluded that scientists make much more discoveries when they AI helps. Now, there are doubts about how well-founded this was, and how much we can learn from how the introduction of AI affects the people who use these tools.