CleanSpark just sold $1.15 billion in zero-coupon convertible notes to buy more power and machinery in its harshest mining environment ever.
The transaction is a 144A private placement expiring in 2032, with an initial conversion price of approximately $19.16, representing a premium of approximately 27.5% to the announced stock price of $15.03.
Approximately $460 million will be allocated directly to the repurchase of CleanSpark stock from bond buyers, with the remainder used to expand its power and land portfolio, build data center infrastructure including AI and high-performance computing capacity, repay a Bitcoin-backed credit line, and cover general corporate expenses.
That single transaction is a cheat sheet of where the minor economy stands in 2025. The terms will reveal who survives, who consolidates, and how much it really costs to remain relevant in a network where the global hash rate has just surpassed 1 zetahash per second.
Whether this bet pays off depends less on the story and more on whether cash flow can support a balance sheet that currently has long-term debt of over $1.7 billion against a treasury of about 12,100 BTC.
Zero percent means something
A zero-coupon convertible bond of this size suggests that credit investors are comfortable being paid in stock options rather than cash interest.
They are betting that despite multiple challenges and price cycles, CleanSpark can remain solvent and maintain sufficient liquidity in its stock for an eventual conversion.
This offers a cost of capital advantage compared to smaller miners, which often rely on high equity dilution and high-yield debt with double-digit coupons. In 2025, only the most efficient miners will be able to borrow such amounts at zero percent. Everyone else is either paying or integrated.
However, the structure comes with risks. This is a leveraged bet on both Bitcoin price and CleanSpark stock performance. If execution stumbles or Bitcoin underperforms, converters become slow-onset dilution bombs.
If the stock trades significantly above $19.16, existing shareholders will be diluted as bondholders convert. Share buybacks further complicate matters because CleanSpark is using $460 million in debt to buy back its stock from the same investors who bought the bonds.
This indicates that management believes the stock is undervalued, but it also means there is less capital available for actual expansion. After the share buyback, approximately $670 million will remain for capital expenditures and debt repayments.
Capital investment and scale in the world of One ZetaHash
The cost of new generation mining rigs and their associated infrastructure typically ranges between $6 million and $10 million per second of capacity.
If CleanSpark were to put all of its incremental capital into mining (which is unlikely given its focus on AI and data centers), that $670 million could fund between 70 and 110 exahashes of additional capacity.
In a network that already has over 1,000 exahashes, even half that would cement CleanSpark as a top-tier hashrate player.
Significant portions are also flowing into building power plants and AI or HPC, and the signs are clear. The minor economy in 2025 is now “grow or be eaten”.
Capital intensity is exploding beyond just purchasing rigs. Miners are building campuses of vertically integrated power and data centers, treating hashing power as part of a broader infrastructure strategy rather than an independent bet on block rewards.
CleanSpark ended its fiscal second quarter with approximately 42.4 exahashes per second and has set a goal of exceeding 50 exahashes by 2025, which at current levels represents approximately 4.9% of the global hash rate.
This increase will allow them to go further, but it also highlights the “treadmill” issue. The network’s hashrate continues to rise, the difficulty adjusts upward, and each exahash produces fewer Bitcoins over time.
Maintaining the status quo after the halving and 1 ZetaHash will require continued reinvestment to maintain revenue per unit of capacity.
Margin stack after halving
CleanSpark’s second-quarter financial numbers show revenue increased 62.5% year-over-year to $181.7 million, but the company had a net loss of $138.8 million and negative adjusted EBITDA. Mining costs will be around $42,700 per Bitcoin, at the end of the efficient curve.
Bitcoin’s price of around $103,000 means that the gross mining margin is around 55% to 60% before selling, general and administrative expenses, interest, hosting, and other expenses.
Energy costs alone accounted for 46% of Bitcoin revenue in the second quarter.
That is the reality after the halving. Block subsidies have been cut in half, network hashrates have reached all-time highs, and hash prices have been compressed to levels that squeeze all but the most efficient operators.
Only miners with cheap and stable power, meaningful scale, and access to low- or zero-coupon capital will be able to maintain positive margins excluding fixed costs.
The 2024 halving did not completely wipe out miners, but rather divided them into two. CleanSpark’s funding shows which side of the divide the company intends to occupy.
Smaller mining companies without fixed power deals or efficient fleets are closing sites, selling assets or raising dilutive equity through market programs.
CleanSpark is doing the opposite, simultaneously conducting stock buybacks and raising debt-like capital, demonstrating confidence that its future hash rate and Bitcoin holdings justify its current stock valuation.
AI side quests: Diversification or narrative sugar?
CleanSpark’s use of proceeds explicitly includes “data center infrastructure” and AI or HPC capabilities. This language reflects a broader industry trend with Core Scientific, Iris Energy, Hut 8, and TeraWulf touting HPC and AI hosting as more profitable uses for their power and infrastructure.
The market is increasingly skeptical of “AI pivot” slides without signed contracts and transparent unit economics.
The framework for determining whether this is true diversification comes down to revenue structure. Will building the AI be done in dollar-denominated, multi-year contracts that de-risk the returns, or will this “maybe someday host the AI” option compete with Bitcoin mining for capital but provide no short-term cash flow?
AI and HPC hosting can generate stable and predictable revenue if contracted properly. However, these dollars compete directly with the value of the increment of Bitcoin mined per megawatt and the option to hold self-mined Bitcoin in the treasury.
Every dollar that CleanSpark spends building AI capabilities is a dollar that is not deployed into hashing power, and the return profile is fundamentally different.
Bitcoin mining provides leveraged exposure to rising Bitcoin prices. AI hosting offers lower volatility and utility-like returns, but also lower upside.
Separate story and cash flow
The current pro forma capital stack includes approximately $640 million in existing debt, $1.15 billion in new convertible debt against equity, and a Bitcoin treasury worth approximately $1.25 billion at $103,000 per Bitcoin.
While the absence of interest expense in the short term improves margins, an overhang for the stock looms if CleanSpark trades well above its conversion price of $19.16.
Return on invested capital plays out in two scenarios. The bullish case rests on Bitcoin remaining above $100,000, hash prices stabilizing, and additional exhash combined with cheap zero percent notes creating strong free cash flow leverage.
A bearish case, on the other hand, would involve a fall in Bitcoin or further compression of the hash price as more hashrate comes online, new capacity yields less, and dilution risks materialize due to lower capitalization.
A rise indicates a state of consolidation phase. Currently, cheap capital and top-quartile power costs are the main moats. Hashing power is becoming institutionalized, with zero percent converters existing alongside large Bitcoin vaults, and the lines between miners and structured Bitcoin funds blurring.
CleanSpark is effectively borrowing against its future mining capacity and Bitcoin holdings, treating the business as an infrastructure-backed financing rather than a speculative venture capital investment.
It’s not about survival capital. That is the cost of entry to be structurally relevant in the world of One Zetahash.
Miners without access to this type of capital are being bought out or shut down. Currently, every dollar has to clear a much higher hurdle than “hashrate increase”. The story is neat and the cash flow tells the true story.
(Tag translation) Bitcoin
