The Hardest Part of This AI Puzzle Was Finding the Question
A new AI benchmark hides the question inside a dossier, separating fact recovery from the ability to spot what needs solving.
Technology
A Tiny Chip Learns Four Ways by Reusing Its Neural Machinery
A small chip adapts to new examples, accumulating keywords, learning from descriptions, and predicting sequences with one compact neural pipeline.
Why a Drone’s Safety Filter Failed at the Moment It Saw Danger
A tiny quadrotor survives wind and motor damage not by spotting danger alone, but by changing its plan while recovery is still possible.
The Robot Improved for 123 Rounds—and Still Missed the Shelf
After 123 rounds, a simulated robot still couldn’t find a fridge’s top shelf. The clues reveal why self-improvement stalled.
The Robot That Builds a Tiny Toy World Before Touching the Real One
ZeroBot builds a virtual object from one view, then trains a robot to handle it in minutes—while revealing the limits of imagined geometry.
When Robot Teams Borrow Evolution’s Rule for Sharing Risk
An evolutionary idea about altruism helps robot teams balance exploration, duplication and risk—and raises the question of who decides which robot matters most.
Why Did Removing a Hard Drive Bring a Sharded Blockchain to a Halt?
A blockchain test uncovered a strange dependency: removing hard-drive delays stopped the chain instead of speeding it up.
The Jigsaw Test That Exposed AI’s Geometric Blind Spot
A carefully engineered jigsaw test finds that vision-language models can recognize images yet falter when many geometric constraints must fit together.
How a Tiny Camera Learns to Send a Hint Instead of an Image
A camera-side encoder turns images into compact machine-readable clues, cutting data traffic and modeled energy without preserving every pixel.
A Robot Learns Housework by Borrowing a Human’s Point of View
A wheeled robot learns whole-body chores from human-worn sensors—but its limits in dexterity, sensing, and skill generalization reveal the price.
The Invisible Charge That Leaves a Tail in a Flash of Light
A thin-film crystal suppresses hidden charge motion, helping an optical switch make cleaner pulses with fewer lingering light tails.
The 'Bad' Superconducting Junction That Made Better Randomness
A misfabricated Josephson junction produced high-entropy random bits at cryogenic temperatures, though speed and testing remain key hurdles.
Can a Magnetic Junction Read a Gyroscope’s Nearly Invisible Twist?
A proposed magnetic gyroscope reads tiny Coriolis motion without capacitive gaps—but its first performance estimates reveal the real puzzle.
A Snake Robot Can Move Its Own Bracing Points Mid-Mission
A new 9 mm continuum robot uses a movable conveyor belt to shift magnetic braces between paired robots and adjust their stiffness in place.
How a Mountain Telescope Reads 63,000 Cold Sensors at Once
At Chile’s Simons Observatory, superconducting circuits let one microwave line read 860 ultra-cold detectors, turning faint sky signals into data.
The Tiny Silicon Channels That Make Extreme-Ultraviolet Light Legible
A patterned silicon membrane uses tiny waveguide channels to reveal the polarization and spectrum of extreme-ultraviolet light.
A Camera Borrowed the Brain’s Reflexes to See Through Motion
A brain-inspired camera system learned to balance motion blur and low light, boosting object recognition while cutting remote AI calls.
The Detector’s Dark-Count Mystery Has More Than One Culprit
A detector’s dark-count jumps can come from its sensor, stray light, or electronics. Afterpulse timing helps reveal which culprit fits.
How a Tiny Bead Taught Hidden Electrodes Exactly Where to Push
A tiny silica bead steers through a crowded microfluidic maze as electrical losses reveal the hidden forces moving it.
A Shaped X-Ray Pulse Could Charge an Iron-Nucleus Battery
A theoretical hard X-ray design stores energy collectively in iron nuclei, but pulse shaping and energy retrieval remain major hurdles.