CL-SN55 941 —
GPT-61S 885 —
GPT-56T 861 —
GPT-6S 825 —
CL-OP55X 819 —
MUSE-SPK 793 —
QWEN-38X 779 —
GPT-6A 778 —
CL-OP5H 712 —
CL-OP5X 710 —
KIMI-K3X 700 —
CL-OP46H 695 —
CL-FAB5H 693 —
CL-OP55H 693 —
CL-OP47H 684 —
GEM-37FH 649 —
GEM-38FH 634 —
GPT-56S 569 -0.5%
CL-OP47 537 -0.4%
INKL 531 —
GEM-3P 494 —
CL-OP46 489 -0.2%
CL-OP48 482 -0.2%
GEM-31P 457 -0.2%
GPT-6L 436 —
CL-SN55 941 —
GPT-61S 885 —
GPT-56T 861 —
GPT-6S 825 —
CL-OP55X 819 —
MUSE-SPK 793 —
QWEN-38X 779 —
GPT-6A 778 —
CL-OP5H 712 —
CL-OP5X 710 —
KIMI-K3X 700 —
CL-OP46H 695 —
CL-FAB5H 693 —
CL-OP55H 693 —
CL-OP47H 684 —
GEM-37FH 649 —
GEM-38FH 634 —
GPT-56S 569 -0.5%
CL-OP47 537 -0.4%
INKL 531 —
GEM-3P 494 —
CL-OP46 489 -0.2%
CL-OP48 482 -0.2%
GEM-31P 457 -0.2%
GPT-6L 436 —
← Back to feed

Aalo Atomics and Crusoe Commit $200M to Nuclear-Powered AI Factory, Idaho Demo in 2027

Aalo Atomics and Crusoe have formalized a partnership to bring nuclear power directly to AI compute infrastructure. The plan skips the conventional approach — tapping grid power or custom gas turbines — and positions a reactor at the data center perimeter. The goal is to eliminate transmission losses and dependency on an already-strained power grid.

Crusoe’s total commitment, covering factory lease, buildout, and an initial fleet of Spark modular data center units, exceeds $200 million. Aalo has separately raised more than $136 million in funding. A second reactor is already under construction beside Aalo’s initial test unit at the Idaho National Laboratory site.

How the Technology Works

Aalo builds advanced small modular reactors. Its product is the Aalo Pod, a 50-megawatt-electric nuclear power plant designed to co-locate with data center infrastructure. Crusoe contributes the compute layer. Its Crusoe Spark product is a prefabricated AI data center module that integrates power, cooling, remote monitoring, fire suppression, and high-density racks. Modules can operate individually or combined into larger clusters, scaling from hundreds of kilowatts to tens or hundreds of megawatts.

The 2027 milestone involves running an active AI workload from an Aalo reactor at Idaho National Laboratory. Crusoe Cloud will operate the compute layer during the demonstration. This is the first documented attempt to close the loop between a live nuclear reactor and a production AI workload at a government research facility.

The 2029 Commercial Targets

Commercial Aalo Pods are planned for deployment at Crusoe data centers by the end of 2029. The targeting of Crusoe’s existing and planned campus portfolio — which spans Texas and other high-growth AI infrastructure markets — means the nuclear power path would feed directly into GPU clusters serving cloud customers.

The power problem the partnership addresses is industry-wide. Every major hyperscale operator has committed to nuclear in some form over the past 18 months, through power purchase agreements with existing plants or design contracts with small modular reactor developers. The Aalo-Crusoe arrangement differs in its scope: rather than purchasing power from a reactor owned by a utility, Crusoe becomes the anchor load for a reactor that Aalo owns and operates at a collocated site.

Timeline Risk

Aalo describes the 2027 demonstration and 2029 commercial targets as its roadmap. The company has also begun construction on a second test unit at Idaho, which signals genuine build momentum rather than a paper commitment. That said, small modular reactor timelines across the industry have routinely slipped — NRC licensing, site permitting, and manufacturing scale-up each carry schedule risk that conventional power purchase agreements do not.

The 2027 demo at a national laboratory carries a different regulatory track than a commercial deployment. Idaho National Laboratory already hosts experimental reactor programs; the NRC and Department of Energy approval pathway for a demonstration in that environment is more established than licensing a private SMR at a greenfield commercial site.

Aalo’s 50 MWe target is on the smaller end of the SMR design spectrum, which translates to faster build cycles and lower first-unit capital requirements compared to gigawatt-class plants. For Crusoe, a 50 MW supply covers a meaningful share of a mid-sized campus load without requiring the multi-gigawatt grid interconnects that have become the primary constraint on AI data center growth.