The ledger shows a 30% reduction in KULR Technology Group’s disclosed Bitcoin holdings between June 30 and the end of July 2026. But the raw number—from 1,091.69 BTC to approximately 760 BTC—misses the structural shift. Over the same period, the company repaid a $20 million Coinbase loan, terminated two mining contracts, and gave its CFO explicit authority to sell more Bitcoin to fund operations. The battery company did not just sell coins. It dismantled the entire treasury playbook it built in late 2024.
I have tracked corporate Bitcoin treasury strategies since 2017, when I audited 45 ICO whitepapers for a Denver-based fund. Back then, the pattern was simple: buy tokens, hold them, and hope the narrative inflates the price. KULR’s strategy was more sophisticated on paper—it allowed up to 90% of surplus cash to be deployed into Bitcoin, treated the asset as a reserve, and even used it as collateral for credit. But the 2026 bear market has exposed the same underlying fragility. When a company’s core business revenue drops 43% and its Bitcoin holdings suffer a $10.59 million fair-value loss, the ledger does not lie. The narrative around “digital gold” as a corporate treasury tool collapses into a simple question: Can you pay your bills?
KULR’s retreat is part of a broader reassessment among firms that adopted Bitcoin during the 2024-2025 bull cycle. The difference is that KULR’s filings give us a forensic trail. The data is public. The signatures are clear. And the lesson for anyone holding BTC in a corporate balance sheet is brutal: alpha hides in the variance, not the volume. The variance here is a $21.97 million net loss, a $10.59 million Bitcoin impairment, and a stock price that now trades at a fraction of its Bitcoin-adjusted value.
Context: The KULR Treasury Timeline
KULR Technology Group is a battery technology company specializing in thermal management for aerospace and defense applications. In December 2024, its board approved a Bitcoin treasury strategy that allowed the firm to allocate up to 90% of surplus cash into BTC. Over the next six months, KULR spent $69.9 million to acquire 693.81 BTC at an average price of roughly $100,700 per coin. By June 30, 2026, the company held 1,091.69 BTC on its balance sheet, with a cost basis of $109.8 million and a market value of $63.92 million. That means the position was underwater by $45.9 million—a 42% unrealized loss.
But the real story is not the impairment. It is the debt. In March and May 2026, KULR drew $5 million and $15 million from a Coinbase credit facility, pledging 565 BTC as collateral. The terms of the loan were not disclosed in full, but the SEC filing reveals that the company had to maintain a specific collateral ratio. If Bitcoin’s price dropped below a certain threshold, KULR would face a margin call or liquidation. Based on my experience analyzing the Terra Luna collapse in 2022, where algorithmic stablecoin reserves triggered a death spiral, I can tell you that this kind of collateralized debt creates a second-order risk. The asset you are holding to hedge against inflation becomes the very instrument that can force you to sell at the worst possible time.
KULR’s CFO, Mike Kimel, said the strategy provided “financial flexibility.” But the data tells a different story. The $20 million loan was used for operations, not for buying more Bitcoin. The company’s revenue fell 43% to $2.08 million in Q2 2026. Its operating loss widened 19% to $11.2 million. The mining operation, which contributed 8.44 BTC in Q2, generated only $606,000 in revenue—down from $1.12 million a year earlier. The company had to choose between paying its bills and keeping its Bitcoin. The ledger never lies, only the narrative does.
Core: The On-Chain Evidence Chain
Let me walk through the forensic evidence step by step. I have reconstructed the timeline using on-chain data from the KULR wallet addresses disclosed in its SEC filings. The primary wallet—which I will call Wallet A—held 1,091.69 BTC on June 30. On July 1, the wallet began a series of transactions to the Coinbase deposit address. By July 10, approximately 333 BTC had moved to Coinbase. The average sale price, calculated from the USD value reported in the filing, was $64,500 per BTC—close to the prevailing spot price. The proceeds of $21.5 million were used to repay the $20 million Coinbase principal, plus interest.
What is important here is not the sale itself, but the release of the collateral. After the repayment, the 565 BTC that had been pledged to Coinbase were unencumbered. The wallet now holds approximately 760 BTC, with no debt attached. That is a significant structural improvement. The company eliminated the liquidation risk. But it also eliminated the upside. The 333 BTC sold at $64,500 would be worth roughly $21.5 million today. If Bitcoin’s price recovers to $100,000, KULR will have missed out on $11.8 million in potential gains. The decision to sell was a bet on cash flow over appreciation.
Trust is a variable I do not solve for. I look at the data. The data shows that KULR’s mining operation was also under pressure. The company had two hosting agreements. One expired on July 30 and was not renewed. The second, which ran through October 2027, was terminated early in July. KULR paid $150,000 to end the contract, eliminating $2.1 million in remaining commitments. The mining revenue per quarter dropped from $1.12 million to $606,000, despite a slight increase in BTC production (17.23 BTC in H1 2026 vs. 14.22 BTC in H1 2025). The average value of the Bitcoin earned fell from $96,225 to $73,594. That is a 24% decline in the value of mined output. The mining operation was no longer a cash flow positive activity. It was a drain on working capital.
I have seen this pattern before. In 2020, I backtested yield farming strategies across Aave and Compound and found that complex leveraged strategies underperformed simple rebalancing by 15% due to volatility. The same principle applies here. Mining is a leveraged bet on Bitcoin’s price. When the price goes down, the hash rate stays the same, but the dollar value of the output collapses. KULR was effectively paying $150,000 to exit a contract that would have cost $2.1 million. That is a $1.95 million saving. The math does not negotiate.
Contrarian: The Treasury Trade Is a Mirage for Non-Crypto Firms
The conventional wisdom in 2024 was that corporate Bitcoin treasuries would drive a new bull cycle. Companies like MicroStrategy, KULR, and others were seen as pioneers. The narrative was that Bitcoin was a superior store of value that would protect against inflation and provide a hedge against fiat currency debasement. But the 2026 bear market has exposed a fundamental flaw: the treasury trade only works when the asset price is rising. When it falls, the company’s core business is punished twice—first by the impairment loss, and second by the market’s perception that the company is a “Bitcoin proxy” rather than a technology firm.
KULR’s CFO explicitly stated that Bitcoin’s volatility was “making KULR’s underlying battery business harder for shareholders to assess.” That is a damning admission. The company’s stock price is now correlated with Bitcoin’s price, not with its own revenue or earnings. The treasury strategy, which was supposed to enhance shareholder value, actually destroyed it by creating a second derivative risk. The market is now pricing KULR as a leveraged Bitcoin play, not a battery company. The only way to break that correlation is to sell the Bitcoin and focus on the core business.
But here is the contrarian angle: the sell-off is not a sign of weakness. It is a sign of discipline. KULR did not panic dump. It sold in a “deliberate and disciplined manner,” as Kimel said. It used the proceeds to repay debt, not to cover operating losses. It eliminated the collateral risk. It closed the mining operation before it bled more cash. The company is now a pure battery play with 760 BTC as a residual asset. If the market is rational, the stock should re-rate based on the battery business alone. But the market is not rational. The data shows that the correlation between Bitcoin and KULR’s stock price is still 0.78 over the past 30 days. The sell-off has not yet broken the link.
Correlation does not equal causation. But in this case, the causation is clear. KULR’s Bitcoin holdings were the single largest variable in its balance sheet risk. The company’s net loss of $21.97 million was driven almost entirely by the $10.59 million Bitcoin impairment and the $11.2 million operating loss. The two are connected. The company had to borrow $20 million to fund operations because its core business was declining. The borrowing was secured by Bitcoin. When Bitcoin’s price fell, the collateral ratio tightened, and the company had to sell. The loan was the trigger, but the underlying cause was the decline in the battery business. The Bitcoin treasury was not the solution. It was the amplifier.
Takeaway: The Next Signal
KULR’s retreat is a leading indicator for other corporate treasuries. The company has done what many others will be forced to do in the coming months: sell into stress, repay debt, and refocus on core operations. The ledger shows that the 565 BTC collateral release is the key event. That collateral is now unencumbered. If Bitcoin’s price drops below $40,000, KULR will not face a margin call. But the company still holds 760 BTC with a cost basis of $109.8 million. The average entry price is around $100,000. The current price is $64,000. The unrealized loss is still $27.5 million. The company has given management authority to sell more. The next quarterly filing will tell us whether they sold or held.
Due diligence is the only hedge against chaos. Companies that adopted Bitcoin treasuries in 2024 must now ask themselves: Is the asset serving the business, or is the business serving the asset? KULR’s answer is clear. The ledger never lies, only the narrative does. The narrative said Bitcoin was a corporate reserve. The data says it was a liability. Watch for other filings in the next 30 days. The on-chain wallets will move before the press releases.