Large language models often lie and cheat. We can’t stop that—but we can make them own up. OpenAI is testing another new way to expose the complicated processes at work inside large language models.

The U.S. Secret Service National Threat Assessment Center (NTAC) has released a new report titled “The State of Behavioral Threat Assessment and Management in K–12 Public Schools.“ This report ...

In classrooms across the country, teachers are facing a crisis that goes beyond test scores and curriculum standards. For many middle and high school educators, managing disruptive behavior has become ...

Each devastating school shooting sparks the instinct to “do something,” notes Jameson Ritter, a certified threat manager and security expert in Minneapolis. But too often, that “something” turns into ...

A facility set to open next week in Greenville is another tool in the state’s expanding approach to breaking harmful cycles related to people experiencing mental health and substance use crises, ...

Anthropic PBC is doubling down on artificial intelligence safety with the release of a new open-source tool that uses AI agents to audit the behavior of large language models. It’s designed to ...

Data shows AI growth is slowing while search usage edges higher – with heavy AI users actually turning to Google search more often. More than 1 in 5 Americans now use AI tools heavily – but ...

Scientists have figured out how to make frozen discs of ice self-propel across a patterned metal surface, according to a new paper published in the journal ACS Applied Materials and Interfaces. It’s ...

Leadership retreats do not typically end in tears. But after 48 hours together over a weekend last fall, a group of emerging leaders from UAB, five Alabama companies and the U.S. Navy were not ready ...

This story was originally published by Chalkbeat. Sign up for their newsletters at ckbe.at/newsletters. Asked to generate intervention plans for struggling students, AI teacher assistants recommended ...