Tether has moved its decentralized search ambitions from concept to a working test, with CEO Paolo Ardoino revealing on Sept. 8, 2026, that the company’s peer-to-peer team stored and searched the full Wikipedia archive across 100 distributed nodes. The demonstration showed a working search interface returning results through the network, with Ardoino describing the performance as “insane” and confirming the team is building what he calls the “Unstoppable Search Engine.”
The project builds on Hypersearch, first disclosed in April 2026, which uses distributed hash table technology and HyperDHT from the Holepunch software stack. Instead of keeping the search index and query process under one centralized operator, the architecture spreads functions among participating nodes. Tether says fault-tolerant structures allow data to self-organize and remain available if individual nodes leave or fail, while reducing dependence on any single server and increasing censorship resistance.
The test environment is still controlled. Tether has not announced a public release date or explained how the engine would crawl and index the open web, manage ranking at scale, or handle spam, manipulated results and conflicting data across a large open network. However, the company is hiring engineers with experience in distributed databases, conflict-free replicated data types, search architecture, ranking algorithms, inverted indexes, NAT traversal and P2P networking, listing HyperDHT, Hyperswarm, Hypercore, Hyperbee, Hyperdrive and UDX among the Holepunch technologies used.
The search project is part of Tether’s broader expansion beyond stablecoins. Tether and Bitfinex previously backed Holepunch and the encrypted communication app Keet, and Tether later applied similar peer-to-peer ideas to its Bitcoin mining tools, including the open-source Mining OS released in February 2026 and an open-source Mining Development Kit released in April. Its earlier Moria mining platform also used Holepunch technology. Tether has also developed QVAC, an open-source toolkit for running AI models locally on consumer devices, with peer-to-peer model distribution through Holepunch. For now, Hypersearch remains under development, with the Wikipedia deployment serving as a large-scale controlled test rather than a public product.
