Artificial Intelligence
Innocent-looking AI reasoning can make bad behavior harder to catch
AI safety monitoring can fail when an AI’s reasoning is the main clue that something has gone wrong, new research suggests.
Science News INDEPENDENT JOURNALISM SINCE 1921
Featured Stories
Trending Stories
We summarize the week's scientific breakthroughs every Thursday.
Spotlight on Back to School
Sponsored Partner Content
Delivering oxygen under pressure: the engineering behind HPO.TECH hyperbaric chamber breathing systems and the BIBS mask
This article has been paid for and developed by HPO.TECH.
Featured Media
NASA’s new space telescope will reveal the universe like never before
NASA’s new Nancy Grace Roman space telescope will tackle nearly every aspect of astrophysics, including dark matter, dark energy and alien planets.
