Bitdeer's $1 billion ATM offering is not a funding round—it's a permission slip to dilute shareholders at will.
Earlier this week, the Bitcoin mining firm filed a prospectus supplement to sell up to $1 billion in Class A shares at market prices. The market reaction was muted, with the stock barely moving. The narrative is simple: Bitdeer needs capital to build its Tydal AI data center in Norway, a project backed by a $1.3 billion letter of credit from JPMorgan Chase. But the structure tells a different story. Ledger logic never lies, only people do. The ledger here is the SEC filing, and it shows a potential 30% dilution for existing shareholders. That's not a funding round—it's a structural risk hidden inside a growth narrative.
Context: The Infrastructure Play Bitdeer is a vertically integrated Bitcoin mining company. It designs its own ASIC chips, operates mining farms, and now plans to pivot into AI/HPC infrastructure. The Tydal project in Norway is the centerpiece: a 100MW AI data center with a target start date of December 2026 for Phase 1 and March 2027 for Phase 2. The project is backed by a $1.3 billion letter of credit from JPMorgan Chase and other financial institutions. But a letter of credit is not cash. It's a conditional promise—if Bitdeer meets construction milestones, the credit can be drawn. If not, the agreement can be terminated. This is a crucial distinction: the project is funded in theory, but not in practice.
Meanwhile, the ATM (at-the-market) offering allows Bitdeer to sell up to $1 billion in shares at any time, at the prevailing market price. The company has already used this tool before, raising $160.7 million from the sale of 9.05 million shares since January 2025. The new filing expands the capacity to $1 billion. Based on the closing price of $10.88 on the filing date, that would require issuing approximately 91.9 million new shares. With 227.4 million shares outstanding, that's a 40% relative dilution—or 28.8% on a fully diluted basis. The math is brutal, but it's not the full story.
Core Analysis: The Structure of Dilution Let me break this down with the same lens I use for DeFi liquidity modeling. In 2020, I built a Python script to track stablecoin liquidity ratios on Uniswap and Aave. The key insight was that when yields become unsustainable, the underlying liquidity is a mirage. Bitdeer's ATM offering is a similar mirage: the market sees a $1 billion funding line and assumes it's earmarked for AI expansion. But the filing explicitly states that proceeds can be used for "working capital, capital expenditures, and other general corporate purposes." Management has broad discretion. The $1 billion is not a dedicated budget for Tydal. It's a funding option that could be used for anything—including plugging operational losses.
The dilution risk is not theoretical; it's encoded in the filing. If the stock price falls, Bitdeer must issue more shares to raise the same amount. At $8 per share, the same $1 billion would require 125 million new shares—a 55% dilution. This creates a negative feedback loop: falling price leads to more issuance, which further depresses price. I've seen this pattern in small-cap tokens during the 2021 bear market. The same dynamics apply here. The difference is that Bitdeer is a regulated equity, not a crypto token. But the economic impact is identical.
Now, let's talk about the letter of credit. A $1.3 billion letter of credit is not a cash infusion. It's a guarantee from JPMorgan that if Bitdeer meets specific milestones, the bank will provide funding. If the project falls behind schedule—and the target start date is still 18 months away—the credit can be revoked. The project is essentially a call option on construction completion. The ATM offering provides the flexibility to raise cash if the credit cannot be drawn. This is a hedge, not a growth catalyst.
Contrarian Angle: The Market Has Misread the Signal The prevailing narrative is bullish: Bitdeer is transforming from a volatile mining company into a stable AI infrastructure provider. The AI pivot is real, and the Tydal project could be a valuable asset. But the contrarian view is that the ATM offering is a signal of overvaluation. Why would management sell shares at market price if they believed the stock was undervalued? The answer is they don't. They are using the stock as currency to fund a project that may not generate returns for years. As I wrote in my CBDC research, CBDCs are infrastructure, not ideology. The same applies here: Tydal is physical infrastructure, but its profitability depends on future AI demand and electricity costs. The market is pricing in a successful outcome, but the pre-mortem analysis reveals multiple failure modes: construction delays, cost overruns, or insufficient demand for AI compute.
The real risk is not the project failing—it's the project succeeding but the financing structure destroying shareholder value. If Bitdeer issues 90 million shares at $10.88 to fund Tydal, and the project eventually generates $100 million in annual EBITDA, the return on equity is less than 5%. That's below the cost of equity. The dilution destroys value even if the project is technically successful. This is a classic capital allocation problem. Based on my experience auditing ICOs in 2017, I've learned that the structure of funding often matters more than the project itself. The ICOs that failed were not necessarily bad ideas; they were bad capital structures. Bitdeer's ATM is a bad capital structure by design.
Takeaway: The Signal Is in the Pace of Issuance For a macro watcher, the key variable is not the Tydal project or the AI narrative. It's the pace of share issuance. Bitdeer has already sold $160 million in shares at an average price of $17.75. The new filing is for $1 billion, but the company may choose to sell only a fraction. The market will price in the dilution gradually, not all at once. The question is whether the stock price will adjust to reflect the new supply. If the stock holds above $10, Bitdeer will likely sell shares aggressively, using the ATM as a primary funding tool. If the stock drops, they will slow down. The signal is not the filing itself; it's the subsequent behavior of management.
Ledger logic never lies, only people do. The Bitdeer ledger shows a company with a promising project but a risky capital structure. The market is still pricing in the promise, not the risk. That disconnect will eventually close. The question is whether you're reading the right ledger—the one that shows the 30% dilution, not the one that shows the AI data center. For now, the smart money is watching the share count, not the narrative.