Antimatter has officially launched as the world’s first vertically integrated neocloud—a new category of AI‑native infrastructure designed to meet the explosive global demand for AI inference. Announced in Cannes, France, the company emerges from the strategic combination of three entities: Data Factory, Policloud, and Hivenet, creating a full‑stack platform that unifies energy sourcing, modular data center hardware, and distributed cloud software into a single AI-optimized ecosystem.
The company enters the market with 1 GW+ of secured power capacity across the US, Europe, and the GCC, and a roadmap to deploy 1,000 distributed micro data centers by 2030. These “Policlouds” are modular, containerized units capable of housing up to 400 GPUs each, deployable in as little as five months—dramatically faster than the 24‑month timelines typical of hyperscale facilities. Antimatter currently operates 10 Policlouds across 8 sites, with 3,400 GPUs already in the field and 26+ MW operational at launch.
CEO David Gurlé emphasized that the bottleneck in AI is no longer intelligence, but energy. “The infrastructure built for the first era of cloud and AI was designed for centralized scale. But the inference era requires a different model: more distributed, faster to deploy, and sovereign by design,” he said.
Antimatter’s model addresses a critical global challenge: the mismatch between soaring AI demand and constrained grid capacity. With global data center demand projected to rise from 55 GW in 2023 to 220 GW by 2030, grid queues and renewable‑energy curtailment have become major barriers. Antimatter’s “energy‑first” approach places compute directly at or near renewable power assets—wind, solar, hydro, and biogas—turning stranded or underutilized energy into productive AI infrastructure.
The company also reports strong early commercial traction, entering the market as a cash‑flow‑positive business with $20 million in annual revenue, $4 million in EBIT, and a diversified customer base spanning energy, public sector, agriculture, and enterprise clients. It targets $250 million+ in revenue within 18 months and $2.5 billion+ by 2030.
Investors from SC Ventures, OneRagtime, Global Ventures, and Inria praised Antimatter’s sovereign‑by‑design architecture, rapid deployment model, and alignment with emerging markets that are leapfrogging legacy infrastructure.
With its integrated stack—energy, hardware, and distributed cloud software—Antimatter positions itself as a foundational player in the AI inference economy, offering infrastructure that is faster, cheaper, more sustainable, and globally distributed.
