AI & ML Nature Is Weird

There’s this 'impossible' crystal structure that lets you squeeze data down as small as you want without it ever breaking.

arXiv · March 17, 2026 · 2603.14999

Roberto Tacconelli

The Takeaway

Most data compression relies on repeating patterns that eventually fail or 'collapse' as you look at different levels of detail. By using the non-repeating Fibonacci patterns found in exotic quasicrystals, scientists discovered they can keep information efficiency perfectly stable at every level, a property that was thought to be mathematically impossible for traditional hierarchies.

From the abstract

We study whether an aperiodic hierarchy can provide a structural advantage for lossless compression over periodic alternatives. We show that Fibonacci quasicrystal tilings avoid the finite-depth collapse that affects periodic hierarchies: usable $n$-gram lookup positions remain non-zero at every level, while periodic tilings collapse after $O(\log p)$ levels for period $p$. This yields an aperiodic hierarchy advantage: dictionary reuse remains available across all scales instead of vanishing bey