“Infrastructure is not a bubble... This isn't a bubble. It's a reflection of a new paradigm that showed no signs of slowing down before AI and now as AI rewrites the rules.” (CEO)
“Not all megawatts are created equal. Our megawatts are strategically located in high-value areas that have multiyear waitlist to secure the power we have today.” (CEO)
“We believe that over time, the companies who’ve allocated and will continue to allocate billions of dollars into compute will be increasingly economically incentivized to pay rising prices in order to deploy their compute faster and with greater certainty.” (CEO)
“Despite being less than 1% of our total development portfolio, we believe that the conversion of just our Moses Lake site to GPU as a service could produce more net operating income per year than we have ever generated with Bitcoin mining.” (CEO)
“Bitfarms is in the strongest financial position in the company's history, and we have a clear vision of how we are going to best utilize this capital to advance our HPC/AI build-outs in North America.” (CFO)
Q&A Batch (1-5 of 9)
Q1 — Michael Colonnese
Topic: HPC economics and mining wind-down
Key points:
Vera Rubin GPU rack density rises from 190 kW (GB300s) to 370+ kW (VR200s); much existing infrastructure incompatible.
GPU margins are very high for newest models, incentivizing faster deployment despite higher infrastructure cost.
LATAM export: Argentina facility shut; Paso Pe (held for sale) represents ~20% of hash rate; expected to pull forward 1–2 years of FCF for reinvestment.
Washington site (~1 exahash) likely offline mid-2026; other sites convert orderly as HPC/AI development proceeds.
Mgmt stance: Bullish — sees clear trend of improving economics due to supply shortage and high GPU margins.
Q2 — Brett Knoblauch
Topic: PA site power expansion and GPU cloud CapEx
Key points:
Panther Creek: ISA-to-ESA conversion and expansion load study received positive indications; Phase 3 could happen quickly (regulatory approval only), Phase 4 likely 2028.
Washington GPU cloud CapEx figure excludes GPUs and some construction costs; GPU manufacturers may fund up to 100% of compute via attractive financing.
Washington expected first fully online, then Sharon; Panther Creek split into phases (2027+). Colocation easier to finance for large PA campuses.
Mgmt stance: Neutral — timing uncertain for PA expansions; prioritizes critical path and affordable cloud scaling.
Q3 — Stephen Glagola
Topic: Washington supply agreement counterparty and GPU cloud vs. colocation
Key points:
$128M critical IT supply agreement is with a large publicly traded American national data center equipment/services company (not T5).
Washington site conversion could be worth “significantly more” than entire Bitcoin mining business over multiple years.
GPU cloud at smaller facility provides learning for hyperscale/enterprise colocation on larger campuses.
Mgmt stance: Bullish — cloud model offers strong FCF foundation and operational knowledge for future large campuses.
Q4 — Mike Grondahl
Topic: Construction bottlenecks and 2026 CapEx guidance
Key points:
Key risk mitigation: strong partners (owners rep, general contractors) and internal project management; specific bottlenecks hard to forecast.
2026 CapEx figures unclear because NVIDIA has not completed Vera Rubin validated reference designs; expected final in Q1 2026.
NVIDIA to produce first Vera Rubin GPUs for own use in Q2 2025; reference design should be final by Q1 2026.
Mgmt stance: Neutral — CapEx guidance deferred until Q1 2026; construction risks managed but unpredictable.
Q5 — Nick Giles
Topic: Future-proofing against rapid rack density increases
Key points:
Rack density evolution: GB200s 150 kW → GB300s 190 kW → Vera Rubin >370 kW; cooling and electrical distribution must change.
NVIDIA considering 800-volt DC systems for Vera Rubin; building for current technology risks obsolescence by completion.
Multiyear agreements (5–15 years) typically lock in infrastructure without material upgrade provisions; best to build for upcoming technology.
Mgmt stance: Cautious — advises spreading facilities across multiple years and building for future tech to mitigate obsolescence risk.