New research shows that AI with human-awareness can accelerate scientific discovery.
New research explores the ways in which artificial intelligence can be used to not only predict but also expand scientific discoveries. Researchers published their research in Nature Human Behaviour. They developed models to predict the inferences of scientists and humans.
They also developed models that avoided the use of human inferences to produce scientifically promising hypotheses. These would likely not be considered until a distant future, or even at all. The authors argue that these two demonstrations – the first allowing the acceleration of human discoveries, and the second identifying and passing over its blindspots – mean that a AI that is aware of humans will allow us to move beyond the current scientific frontier.
James A. Evans is the Max Palevsky professor in the Department of Sociology, and the director of Knowledge Lab. He says that by increasing awareness of what people do, it will be possible to improve predictions and to leapfrog ahead of them, accelerating science. You can also determine what people are unable to do at the moment, or will not be able do in decades or longer. You can enhance them by giving them this kind of complementary intelligence.\”
Source:
https://techxplore.com/news/2023-07-human-aware-ai-scientific-discoveries.html
