A cryptographic trick uses RAM speed to make a powerful GPU run slower than a cheap laptop.
April 20, 2026
Original Paper
PoSME: Proof of Sequential Memory Execution via Latency-Bound Pointer Chasing with Causal Hash Binding
arXiv · 2604.15751
The Takeaway
PoSME forces a computer to follow a sequential path of memory access that cannot be accelerated by parallel processing. By tying the computation to hardware latency, the system ensures that more hardware does not mean faster results. This makes massive GPU clusters 14 to 19 times slower than a single CPU for these specific tasks. It creates a verifiable clock that proves exactly how much time has passed since a computation started. This technology prevents wealthy actors from using massive hardware to cheat time-based security protocols.
From the abstract
We introduce PoSME (Proof of Sequential Memory Execution), a cryptographic primitive that enforces sustained sequential computation via latency-bound pointer chasing over a mutable arena. Each step reads data-dependent addresses, writes a block whose value and causal hash are mutually dependent (symbiotic binding), and chains the result into a global transcript. This yields three properties: (1) strict linear sequential memory-step enforcement, (2) high time-memory trade-off resistance (a tenfol