Researchers are proposing that a modernized version of an experiment older than the United States could become the most sensitive dark matter detector ever built. The Cavendish torsion balance, originally designed to weigh the Earth by measuring gravitational attraction between lead spheres, might be repurposed to hunt for axions (hypothetical particles that could explain what dark matter actually is). The core idea exploits resonance: if you tune a suspended pendulum to oscillate at exactly the frequency an axion field would produce as it streams through our planet, the signal gets amplified to detectable levels. Same physics that shatters wine glasses, applied to one of cosmology's hardest problems. The timing matters because conventional dark matter searches are hitting a wall. ADMX (Axion Dark Matter eXperiment) at the University of Washington uses superconducting magnets and microwave cavities cooled to near absolute zero, converting hypothetical axions into photons inside a strong magnetic field. These setups cost tens of millions to build and require constant cryogenic maintenance. After decades of operation, they've scanned only narrow slices of the axion mass range with no confirmed detections. The torsion balance approach targets the same particles but through gravitational coupling instead of electromagnetic conversion, a completely different detection channel that might work where cavity experiments have come up empty. The claimed sensitivity advantage (10,000 times better than existing instruments in certain mass ranges) sounds extraordinary, but it comes with massive asterisks. That projection assumes you can isolate the pendulum from every imaginable source of mechanical noise. A truck driving half a mile away creates vibrations that couple into the suspension fiber. Thermal expansion of the laboratory walls as temperature drifts by fractions of a degree produces torque signals larger than what an axion would generate. Even gravitational pull from people walking near the apparatus creates detectable effects. The original Cavendish experiment took weeks to get a single measurement because environmental noise was so dominant. A dark matter version would need isolation orders of magnitude better. What makes this proposal interesting is the cost-benefit calculation. Current axion detectors require infrastructure you can't buy off a shelf: custom superconducting magnets, dilution refrigerators, kilometers of specialized cabling. A torsion balance needs precision machining, vacuum chambers, and vibration isolation (expensive, but within reach of a university physics department rather than a national laboratory consortium). If the sensitivity claims hold up in a working prototype, you could deploy multiple instruments tuned to different axion mass ranges for less than the cost of one ADMX-scale facility. That's the kind of economics that gets funding committees interested, especially after years of null results from more expensive approaches. The physics community has been here before with bold sensitivity claims that evaporate once you build the actual detector. Thermal noise sets a hard floor: at any temperature above absolute zero, atoms vibrate randomly, and those vibrations swamp faint signals. Cooling a macroscopic mechanical system to cryogenic temperatures introduces new problems (differential contraction of materials, boil-off of cryogens, electromagnetic interference from pumps and heaters). The proposal doesn't yet describe how to overcome these practical obstacles. No prototype has been reported, no experimental validation of the projected sensitivity exists. It's a theoretical exercise at this stage, not engineering reality.