Nscale and Anthropic in talks for $45 billion compute deal

By Splitfeed EditorialLast updated Oct 19, 2026, 20:50 UTC

On October 19, 2026, Bloomberg News reported that Anthropic is in talks with cloud computing provider Nscale for a deal worth up to $45 billion. According to Traders Union, the infrastructure tied to the agreement is linked to West Virginia. Seeking Alpha described the potential Nscale deal as Anthropic's second agreement at this scale, suggesting a prior commitment of similar size had already been arranged or was in progress. Separately, 36Kr reported that Anthropic recently closed a funding round that placed its valuation at $320.8 billion. Neither Anthropic nor Nscale has issued a confirmed statement, and the talks have not been described as concluded.

The scale of what is being negotiated here is worth pausing on. A single compute deal worth up to $45 billion would represent an extraordinary concentration of capital around one AI company's infrastructure needs, and the reported second agreement of this size in quick succession signals that Anthropic is planning for compute requirements far beyond what its current systems demand. The West Virginia location matters in a practical sense, as data center development in that region is often tied to energy availability and land costs, both of which are increasingly contested variables in the AI infrastructure build-out. What remains genuinely in dispute is whether commitments of this magnitude reflect sound capital allocation or a speculative overshoot driven by competitive pressure among frontier AI labs.

No reactions from the Pro-AI, Anti-AI, or Middle Ground camps had been published at the time of writing. Pro-AI commentators would typically welcome news like this as evidence that serious investors and infrastructure partners view frontier AI development as a long-term bet worth underwriting at enormous scale. Anti-AI voices would likely raise concerns about the concentration of resources, the environmental footprint of large data center buildouts, and whether this level of spending accelerates AI development faster than safety or regulatory frameworks can keep pace. Middle Ground observers would probably focus on the financial risk involved in multi-decade infrastructure commitments and whether the projected demand for compute will materialize as expected.

The clearest near-term signals to watch are any official confirmation from Anthropic or Nscale that a deal has been signed, disclosures about the specific terms or duration of the agreement, and any regulatory or permitting activity tied to data center construction in West Virginia that would indicate how far along the project actually is.