AI
AI Research Automation Needs More Than a Task Percentage
Anthropic’s new automation measurements and OpenAI’s incident disclosures show why delegated work, monitoring and research outcomes need separate…
Section
Future
Reporting, analysis, interviews, and ideas from the TENS Magazine newsroom.
From the newsroom
Stories
Longevity
LongevityBench Opens Aging AI to a Reproducibility Test
An open aging-AI toolkit makes research more inspectable. Its scores need task-specific context, reproducible methods and clear limits…
AI
AI Classroom Simulations Need Evidence Beyond Working Software
Google’s learning interactives bring separate tests for software quality, teacher review effort and student understanding into focus.
Longevity
Aging-Rate Indicators Face a Test of Which Experiments They Choose
A new mouse-research framework could help prioritize longevity experiments. Its value will depend on prospective testing, missed candidates…
AI
Faster AI Search Makes Result Quality a Product Decision
Google’s Retrieve-for-Train research shows why faster AI search needs separate tests for useful results, diversity and changing catalogs.
Longevity
Sex-Specific Aging Clocks Put the Reference Population First
A new study makes reference populations central to aging-clock interpretation. TENS proposes clearer reporting of model versions, biomarker…
AI
HardFlow Separates AI Planning Constraints From Deployment Proof
MIT’s HardFlow research shows why AI planning results need separate evidence for generated routes, operating assumptions and real-world…
Longevity
A Brain Microprotein Atlas Makes the Case for Better Evidence Labels
A new brain microprotein atlas shows why protein detection, cellular function and patient benefit need separate evidence labels…
AI
BigQuery’s AI Analytics Make Statistical Labels a Product Issue
Google’s new analytics functions put the interpretation of statistical outputs at the center of reliable AI investigations.