Revealing

• The CDC posted its first large set of genetic data from 1,000 measles viruses circulating in the US last year, with more data expected soon to assess if elimination status has been lost. • In 2025, measles cases reached 2,285 across 44 states—the worst in over 30 years—while 2026 has seen 1,575 cases by late March amid declining vaccination rates. • Delays in data release stemmed from mass layoffs and resignations; researchers criticize the Trump administration's budget cuts and slow outbreak response.

• Engineers at Washington University in St. Louis discovered that cells can sense far beyond the surfaces they touch through a process called "depth mechano-sensing," with clusters of normal epithelial cells detecting layers as far as 100 microns away by tugging on surrounding collagen fibers. • Individual cancer cells can probe about 10 microns ahead, while normal cell clusters working together can sense 10 times farther, helping cells decide where to migrate and providing new insights into cancer cell behavior. • The findings, published in PNAS, could reveal new targets to prevent cancer from spreading and advance understanding of how cancer cells move through the body.

Tommy Thompson refused to give up the location of 500 missing coins found in a historic shipwreckA US treasure hunter who was imprisoned for 10 years after refusing to reveal the location of missing gold coins has been released from prison, without officials apparently ever learning where that gold is.Tommy Thompson – a renowned salvager who in 1998 found the long-lost, so-called Ship of Gold near South Carolina – was freed from federal prison on 4 March, records and reports recently indicated. Continue reading...

Researchers at Texas A&M University have developed what they describe as the most challenging AI benchmark test to date, with results that contradict earlier expectations about artificial intelligence capabilities. The comprehensive study involved a large team of scientists investigating how cutting-edge AI systems perform on extreme difficulty assessments as these models become increasingly sophisticated. The findings suggest that current AI systems are achieving unexpected performance levels on this rigorous evaluation framework, providing important insights into the true capabilities and limitations of advanced machine learning models.

Scientists at Texas A&M University created the most challenging AI test to date, involving over 50 collaborators, to evaluate advanced language models on complex reasoning tasks. Initial results showed even top AI systems struggling significantly, highlighting gaps in current capabilities despite acing simpler benchmarks. The benchmark, detailed in a peer-reviewed paper, pushes boundaries in AI safety and robustness research. Future iterations will incorporate multimodal challenges to further stress-test emerging models.