Officials in Oakland sharply increased the number of homeless encampments they cleared in the months after the 2024 U.S. Supreme Court decision made it easier for municipalities nationwide to do so, new research from UC Berkeley shows. Many sites have been closed repeatedly, the analysis found — one was swept 18 times in a four-year period. More recently, efforts to close camps have shifted into census tracts that have higher poverty rates and larger shares of Hispanic residents. 

Samuel Kruger, associate professor of finance; John Griffin, James A. Elkins Centennial Chair in Finance; and doctoral student Prateek Mahajan found that fraud in government-funded pandemic loans explained 22.5% of the average increase in housing prices during 2020 and 2021.

 

In new research, Yan Leng, assistant professor of information, risk, and operations management at the McCombs School of Business at The University of Texas at Austin, has devised a sort of personality test — more precisely, a behavioral audit — for large language models (LLMs), the technology that drives products such as ChatGPT.

By understanding an LLM’s existing tendencies, an organization can decide whether an available model already fits its values and usage scenarios. If not, it might need to fine-tune a model before putting it to use.

Leng compares her framework to trying to understand a person through their actions and thought processes. “For a human, we would have our values, and our values would dictate how we make decisions, so we try to have that for LLMs as well,” she says.

A new study from the Child Mind Institute finds that negative online experiences are common among youth with mental health and neurodevelopmental conditions, and that most incidents go unreported through platform tools. Published in JAACAP Open, the study surveyed 1,009 youth ages 9 to 15 and found that more than one in four experienced at least one negative online incident in the past year, including cyberbullying, harassment, and other forms of digital harm. Among those who had such an experience, nearly 69% reported multiple incidents, yet only 20% used platform reporting tools. Researchers identified key barriers to reporting, including confusion about how to report, uncertainty about what qualifies, and emotional barriers such as embarrassment and fear of consequences. The findings suggest that online safety efforts need to be tailored to youth who may already be managing mental health or developmental challenges, and that reporting systems should be clearer, safer, and easier for young people to use.