The Mining-to-AI Infrastructure Pivot:
Why It Matters

In 2024, the dominant narrative in Bitcoin mining was halving math — how operators would survive the revenue compression when block rewards dropped from 6.25 to 3.125 BTC. By mid-2026, that narrative has been completely supplanted. Public miners aren't talking about survival. They're talking about AI.

$70B+ in AI/HPC contracts signed by public miners in 2026

The numbers are hard to overstate. Core Scientific signed a $10.2 billion agreement with CoreWeave to convert a significant portion of its mining infrastructure to AI colocation. Hut 8 announced a $7 billion Google-backed AI infrastructure lease. TeraWulf has committed 72 megawatts to AI/HPC hosting at its nuclear-powered Lake Mariner facility. Cipher Mining, Marathon Digital, and Bitfarms are all in various stages of similar pivots.

In February 2026, Morgan Stanley initiated coverage of Bitcoin miners with an explicit framing: not as cryptocurrency companies, but as energy infrastructure for the AI economy. The thesis is straightforward. Miners have done the hard work — they've found power, built grid interconnections, erected facilities, and established operational track records with utilities. Those assets are exactly what the AI data center build-out needs, and hyperscalers are paying premium prices to access them quickly.

Why the pivot makes economic sense

Bitcoin mining generates approximately $80–150 per kilowatt per month in revenue, depending on BTC price and network difficulty. AI/HPC colocation generates $150–300 per kilowatt per month under multi-year contracts — with far less volatility, since it's tied to long-term agreements rather than daily BTC price movements.

For operators with power-ready, fiber-connected facilities, the economics of shifting capacity toward AI workloads are compelling. Mining remains a valuable use of power during periods of lower AI demand — it fills otherwise idle capacity with revenue-generating work. The optimal model is hybrid: AI/HPC as the anchor, mining as dynamic fill load.

What this means for Pursuit Link

Pursuit Link is a smaller operator, but we're executing the same strategic playbook — intentionally and ahead of scale. Our TVA power infrastructure and EPB dark fiber connectivity are the exact profile that hyperscalers, GPU cloud providers, and AI training companies are seeking. Our NVIDIA infrastructure positions us to serve inference workloads locally, while our existing ASIC hosting operation provides cash-flowing infrastructure from day one.

The mining-to-AI transition isn't a story about miners becoming something else. It's a story about power-rich, fiber-connected operators realizing their underlying asset — reliable, affordable compute capacity — is the scarcest resource in the AI economy. We've been building that asset in Chattanooga since inception.

The same TVA rate agreement that makes our Bitcoin hosting margins compelling makes our AI infrastructure costs competitive with any primary data center market. The EPB dark fiber that connects us at up to 400 Gbps gives us the latency profile that AI inference and model serving require. The self-healing smart grid that EPB built — 1,100+ switches, seconds to restoration — gives us the uptime reliability that institutional AI tenants demand.

The opportunity for investors and anchor tenants

The data center sector is entering a generational capital deployment cycle. An estimated $3 trillion in infrastructure investment will be required by 2030 to meet AI compute demand. Primary markets — Northern Virginia, Phoenix, Chicago, Dallas — face 4+ year grid connection wait times and construction costs exceeding $11M per megawatt. Operators who can deliver power-ready capacity now, in markets with structural cost advantages, are commanding premium valuations and long-term lease commitments.

Pursuit Link is positioned to be that operator in the Tennessee Valley region. If you're evaluating AI infrastructure opportunities or looking for anchor compute capacity in a cost-advantaged market, we'd like to have that conversation.