2023-07-14 15:31:15
by Elena Giovanna Bignami
The scientific journal Science dedicates a monograph to the hottest topic of the moment. Many questions but no answers. Just a contribution to better understand
There have been tremendous advances in the development and application of Artificial Intelligence (AI) to science and society. But will AI eclipse humans, or will we find a way to collaborate safely and fairly, allowing us to go further?
To these and other questions it tries to give not answers but contributions to help us better understand the special issue that Science wanted to dedicate to the hottest topic of the moment. Some authors in various scientific articles and in various fields underline the progress of new technologies and in particular of Artificial Intelligence (AI) evaluating their field of application, the possible diffusion and also technical, legal and ethical problems.
Racial variables
Some studies have shown that AI models can infer racial variables, albeit in broad and simple categories, directly from medical images (such as chest X-rays and cardiac ultrasounds), despite no known specific racial correlates in clinical diagnostic images.
James Zou and colleagues say that “AI’s ability to predict racial variables from medical images could be useful for monitoring health disparities and ensuring that the algorithms perform well in different populations.” Matthew DeCamp and Charlotta Lindvall point out how assessing variables that use AI in health care has tended to remove bias from datasets, analyzes or in AI development teams. However, to reduce or eliminate bias in AI we need to remove selection variables in the numbers fed to AI algorithms.
Machine learning to decipher how animals communicate
In another study, Rutz and collaborators argue that Machine Learning-ML (a branch of AI) could be used to decipher animal communication systems with consequent potential benefits in animal welfare.
Controlled game systems
In the fourth contribution, attention is placed on how AI can further investigate its application in controlled gaming systems, to improve «Problem Solving» skills, which can then be applied in the real world with subsequent benefit, and that is, as a primordial simulation.
Generative AI and copyright
During a comparison related to the use of «generative AI», Pamela Samuelson underlines how there may be substantial variations in the future in the dissemination and therefore in the use of AI following various legal actions by artists related to copyright due to the more and more development and diffusion of this type of AI, such as the application that generates video, images, audio and texts.
Economic inequalities
Some economics scholars also argue that AI systems can increase inequalities between people in this field. Ajay Agrawal and collaborators, however, focus attention on the consequences of the application of AI in the labor market by hypothesizing a reduction of the ever-increasing gap between the different social and economic conditions of different citizens. In fact, they think that by creating codified and easy-to-use models, even less skilled workers could begin to carry out activities that previously required specialized training, and therefore not easily accessible to everyone.
The role in the field of infectious diseases
In a review, Wong underlines how AI and ML have given a significant boost to research in the infectious disease field, leading to an ever more complete and profound understanding of the pathological mechanisms of infections and therefore the development of new drugs.
Chemicals management
Bing Huang, on the other hand, focuses on the crucial role that ML could have as software capable of regularly handling exotic chemicals and formulations within laboratories with relative autonomy, calling it «Functional Theory of Fate» – fundamental in the science of chemistry and of materials due to its relatively high predictive power.
I robot in medicina
Also, there are different interpretations of using AI for
sophisticated robots in the medical field, which could carry out precise and real-time diagnostic images and therefore assist clinicians during surgical interventions.
Prosthesis and rehabilitation
Finally, the use of AI in the prosthetic-rehabilitation field could generate more and more advanced aids so as to create personalized medicine and assistance, until reaching AI-based prostheses that work uniformly with humans.
*Professor of Anesthesia and Intensive Care, University of Parma; Artificial Intelligence expert of the Italian Society of Anesthesia Analgesia Resuscitation and Intensive Care
July 14, 2023 (change July 14, 2023 | 3:30 pm)
© Time.News
#artificial #intelligence #eclipse #humans #find #collaborate #safely #time.news