KULR Technology Group, Inc. (KULR) stock intrinsic value fair value estimate & owner earnings analysis

KULR Technology Group designs battery systems and thermal-safety products (thermal runaway shields, phase-change heat sinks, safety cases, internal short-circuit battery cells and devices) for autonomous platforms, digital infrastructure, e-mobility and space, and is establishing a battery production facility in Houston, Texas. Since 2025 it has also held bitcoin as a treasury asset and leases mining machines that earn bitcoin. This is the first time we have analysed the company. What the business earns. In every year we reviewed (2016 through 2025) the company consumed cash in its operations rather than producing it. On our measure of owner earnings (operating cash flow, less working-capital swings, less stock compensation, less the equipment spending needed just to stay in business) 2025 was a loss of about $45.0 million, or $1.13 per share, on 39.7 million diluted shares. The prior years, adjusted for the June 2025 one-for-eight reverse split, were: 2020 -$0.23, 2021 -$0.96, 2022 -$1.56, 2023 -$1.38, 2024 -$0.58. We use 2021 to 2025 as the window, since 2021 is the first year revenue exceeded $2 million and earlier years describe a much smaller business; the median of that window is -$1.13 per share. The loss per share got worse in 2025 (from -$0.58 to -$1.13), and our base case starts year one at that same level of loss rather than assuming any improvement, because nothing dated in the filings shows the trend reversing. We therefore give the operating business no value from future earnings. Why the 2025 revenue growth does not change that. Revenue rose 51% to $16.2 million, but $7.0 million of it was bitcoin mining, which ran at a negative gross margin (cost of $7.5 million against $7.0 million of revenue). Without mining, revenue fell from $10.7 million to $9.1 million, a drop of about 15%, because a $2.7 million technology license recognised in 2024 did not recur and contract-services revenue halved. Total gross profit was only $0.8 million, about 5% of revenue. In the first half of 2026, revenue was $6.0 million against $6.1 million a year earlier, and operating cash outflow was $17.7 million against $21.5 million, a smaller burn but still a large one. The company also wrote off its investment and receivables in a German customer that filed for insolvency in November 2025 (about $3.3 million equity stake, $2.1 million loan and $0.8 million receivable) and wrote its exoskeleton inventory down to zero. What is left: the treasury. The value the shares can rest on is the balance sheet. At June 30, 2026 the company held bitcoin worth about $63.9 million (it bought 1,000.99 coins at an average of about $100,600 and the price was about $66,000 in late March 2026), cash of $12.8 million, and shareholders' equity of $72.6 million, or about $1.57 per share on 46.3 million shares. It had a $20.0 million bitcoin-backed loan, which it said it repaid after quarter-end using bitcoin proceeds. It raised $123.2 million by selling stock in 2025, which lifted the share count from 33.1 million at the end of 2024 to 46.0 million at the end of 2025, and paused that program through June 30, 2026. Management states that cash and bitcoin are sufficient for the next twelve months. The risk is the pace at which operations consume that pool: at 2025 levels, cash burn of about $36.6 million a year before working-capital changes would use roughly $70 million in two years. Bear case, $0.00 per share. Operations keep consuming cash at the 2025 rate, bitcoin stays near its current price, and the company must keep selling coins or shares to fund itself. The treasury is spent and diluted away before the business reaches break-even, leaving nothing for shareholders. This is not below zero because the borrowings are small next to the treasury. Base case, $0.05 per share. We credit no growth, since nothing is contracted or proven repeatable at a profitable scale (the Texas Space Commission grant, awarded September 23, 2025, only reimburses costs, and the ten-year CF Cathode licence in Japan has a total contract value of only $1.8 million, of which $1.7 million was recognised in 2024 and none in 2025). The operating business continues to burn about $35 million a year for two years, the pool of about $72.6 million of equity is drawn down to roughly $2.6 million, or about $0.05 per share, and nothing further is credited. Bull case, $0.80 per share. The company cuts operating losses so that it burns only one more year at about $35 million, then stops consuming cash. That leaves about $37.6 million of the current $72.6 million equity, or about $0.80 per share on 46.3 million shares. The first-half 2026 improvement in cash burn (from $21.5 million to $17.7 million) is the only evidence toward this outcome, so we do not go beyond it. Bitcoin price recovery is not credited to any case, and the bitcoin is counted at its June 30, 2026 fair value. Judgements worth knowing. Stock compensation ($6.5 million in 2025) is treated as a real cost. Equipment spending was split between upkeep and growth using the six-year average of about 19% of revenue; in 2025 about $1.1 million of the $3.0 million was treated as growth. The bitcoin treasury is not part of the earnings stream, so we count it only as the pool that funds the losses, not as a separate addition. We could not reconcile the jump in total liabilities from $7.4 million at the end of 2025 to $27.5 million at June 30, 2026 in detail, but it coincides with the $20.0 million bitcoin-backed loan. Margin-of-safety price, $0.55 per share, applicable only if the operating losses stop. At today's cash burn no margin-of-safety price applies, because a share is a claim on a shrinking pool. If burn falls to about $10 million a year or less, the shares become mostly a claim on the roughly $1.57 per share of equity, and the margin-of-safety price is set more than 60% below that for a small company whose main asset is volatile bitcoin. What would make this wrong: a large, contracted increase in profitable battery or thermal-product orders (not evident today), or a much higher bitcoin price, would raise value; faster burn, more share issuance, or a further fall in bitcoin would lower it to zero.