SHELFCRITTER AI Digest

COMPANY

Naver

companytopic-note

Overview

Naver is South Korea’s largest internet company and the operator of the HyperCLOVA X family of large language models. In June 2026, Naver announced a multi-year sovereign-AI roadmap with NVIDIA to build AI factories on NVIDIA’s DSX platform, joining the Nemotron Coalition and fine-tuning open Nemotron models into the next generation of HyperCLOVA X. The buildout pairs hyperscaler-grade capex on US silicon with a domestic-distribution sovereign-AI play, distinct from the state-procurement model the UK is running in parallel.

Timeline

  • 2026-06-08-AI-Digest — Naver and NVIDIA road map AI factories on NVIDIA’s DSX platform: 55 MW operational from H1 2027, scaling to ~200 MW by 2028 and a long-term path toward gigawatt scale. Naver joins the Nemotron Coalition as the first Korean member, fine-tunes open Nemotron models into next-gen HyperCLOVA X, and adds a “Seoul World Model” on NVIDIA Cosmos for agentic services. Bloomberg covers the announcement alongside the same-day UK AI Hardware Plan procurement story — the corpus distinction is mechanism (Naver’s lever is hyperscaler capex on US silicon; the UK’s lever is industrial policy buying domestic chips), and the corpus’s standing risk is conflating the two under a single “sovereign AI” headline. The number to carry forward is 55 MW as the first step toward gigawatt, not the gigawatt itself, and the operational date is 2027 — load-bearing for anyone modelling Korean inference capacity into 2028.

Key Developments

  1. Naver–NVIDIA DSX Roadmap (June 8, 2026): 55 MW operational from H1 2027 scaling to ~200 MW by 2028 with a long-term gigawatt-track buildout; the announcement formalises Naver as the anchor Korean buyer for NVIDIA’s DSX platform and pairs hyperscaler-grade capex with sovereign-AI distribution. The 55 MW number is the calibration: this is the first step on a multi-year arc, not the gigawatt itself.

  2. Nemotron Coalition Membership + HyperCLOVA X Fine-Tunes: First Korean member of NVIDIA’s Nemotron Coalition, with the stated plan to fine-tune open Nemotron models into the next generation of HyperCLOVA X. Extends Nemotron’s coalition footprint into the Korean sovereign-AI lane and positions HyperCLOVA X as the consumer-distribution surface for a Nemotron-derived base.

  3. Seoul World Model on NVIDIA Cosmos: A location-grounded world model for agentic services built on NVIDIA Cosmos — paired with the DSX capacity buildout, the play is to own both the compute substrate and a domestic agent-training environment without ceding either to a hyperscaler counterparty.