New AI framework aims to remove bias in key areas such as health, education, and recruitment
Peer-Reviewed Publication
Updates every hour. Last Updated: 16-Jul-2025 20:11 ET (17-Jul-2025 00:11 GMT/UTC)
Researchers from the University of Navarra's Data Science and Artificial Intelligence Institute (DATAI) have developed a new AI framework to reduce bias in critical decision-making areas such as health, education, and recruitment. Their methodology optimizes machine learning models to ensure fairness by addressing inequalities related to race, gender, and socioeconomic status, among other possible algorithmic discriminations. Published in Machine Learning, the study combines conformal prediction techniques with evolutionary learning to achieve reliable and unbiased AI predictions. The researchers tested their approach on real-world datasets, demonstrating that it reduces discrimination without compromising accuracy. Their work provides policymakers and businesses with AI models that balance efficiency and fairness, aligning with ethical AI principles and legal requirements. The team has publicly made their code and data available to promote transparency and further research in responsible AI development.
A new report issued by the Purdue Applied Research Institute’s Digital Innovation in Agri-Food Systems Laboratory, the research arm of DIAL Ventures, offers multiple strategies to help the agrifood sector navigate climate change-related challenges throughout the agricultural value chain.
“The findings of the Climate-Smart Agrifood Opportunities report provide a road map for stakeholders across the agrifood sector to identify and act on opportunities for sustainability,” said DIAL Ventures research manager Lourival Monaco, research assistant professor in agricultural economics. “By addressing the challenges collaboratively, the sector can build resilience and adapt to the evolving demands
University of Chicago researchers analyzed data to identify trends that could shed light on why Black women with triple-negative breast cancer (TNBC) have worse survival rates. Since new immunotherapy treatments for metastatic and early-stage TNBC were approved in 2019 and 2021, Howard and Freeman also examined disparities in access to these new treatments. While they found that much of the inequality they observed between racial and ethnic groups could be explained by socioeconomic disparities, they also found that Black women with TNBC were less likely to receive immunotherapy treatment even when accounting for those factors.
A multi-institutional research team led by Osaka University has introduced a piezoelectric device that can measure acceleration and pressure simultaneously. The device can be manufactured at room temperature and is made from inexpensive materials, which makes it a promising candidate for a wide range of applications, including the maintenance of industrial machinery and disaster mitigation.
The addition of this year's members brings the total of National Academy members in the College of Engineering to 38.