Stark Power Buys SAGE and Moves the Risk to Project Funding
The $50 million SAGE acquisition gives Stark Power a US data-center platform with defined projects, land options and a Powered Land timetable. The value is no longer only a shell-company story, but it now depends on closing the offering, option exercises and the conversion of land, power and customers into cash.
STARK POWER-M signed on May 29, through its wholly controlled SunSpear subsidiary, an agreement to acquire 100% of SAGE for $50 million in cash. This is stronger than a memorandum of understanding or a headline about entering data centers, because there is now a purchase agreement, a project list, a Powered Land definition (land rights, material permits and a basic power solution), and a first-phase timetable for 1.55 GW IT in 2027-2029. The economics are more complicated than the entry price: the offering proceeds are intended to finance the acquisition, but SAGE itself still has no revenue-generating business, no customers and no order backlog. The move also replaces a route in which Sagebrush 1 might have been sold through milestone payments with a route in which the company holds development risk for longer in an effort to capture more value. Shareholders receive a more concrete exposure to a US infrastructure portfolio and pay for it through dilution, reliance on option exercises and dependence on land, grid connection, permits and customers. The next read depends on closing the transaction, receiving the offering proceeds, short-dated option exercises and progress in at least one project from land under option to an asset that a buyer is willing to pay for.
A Defined Portfolio Arrives Without Near-Term Revenue
The May 31 acquisition changes the story of Stark Power from a general entry into US infrastructure into a transaction with an identifiable asset. SAGE is developing five data-center campuses with aggregate capacity of about 5.55 GW IT, including first phases of about 1.55 GW IT expected to reach Powered Land during 2027-2029. It is also advancing three gas-fired co-located power plants with aggregate capacity of about 1.55 GW.
The important gap sits inside the term Powered Land. This is not an operating data center, and it is not a contract with a cloud customer. It is a stage in which there are land rights, material permits and a power solution that enables a buyer to continue toward construction. SAGE generally plans to sell a full campus after the first phase reaches this stage, or sell phases separately when that is economically preferable. The acquisition therefore does not bring an operating business into Stark Power. It gives the company development rights whose value depends on operational progress.
SAGE's numbers reinforce that distinction. At the end of 2025 it had about $2.17 million in cash and about $2.31 million in net current assets, and it recorded a loss of about $420 thousand since its formation in August. It had no material customers, no order backlog, no fully owned land assets and no material financing arrangements with suppliers or customers. That does not invalidate the deal, but it defines it: Stark Power is buying a development platform, not an operating business.
Dropping the Sagebrush 1 Sale Route Shifts Value Into Development
The detail that sharpens the transaction is not only the purchase price. On March 24, the company described a possible sale of Sagebrush 1 for milestone payments of about $45 million and profit participation of up to another $65 million, or up to about $110 million in aggregate. After the SAGE acquisition agreement, the parties chose not to proceed with that route, based on the view that independent development of Sagebrush 1 to Powered Land could create higher value.
That decision raises the theoretical upside and changes the risk profile. Instead of a route that might have brought staged consideration from one project, the company carries more of the land, planning, power and final-buyer risk for longer. Sagebrush 1 is also the most advanced project in the portfolio: it is in the SPP power market, it benefits from surrounding infrastructure, and a nearby local power plant is expected to reach commercial operation in 2029. The first phase is still expected to reach Powered Land only in 2028, with later phases extending to 2032.
For investors, the point is not simply "more projects." It is more control over monetization timing. If the company advances Sagebrush 1 itself, potential consideration could exceed the abandoned sale route. If progress is delayed, the cash path gets longer just when the company must prove that the shift from shell company to infrastructure platform is not supported only by options.
The Options Finance the Entry and Increase Dilution
On May 27, Stark Power received classified-investor commitments for an amended offering structure. The May 31 shelf offering report sets the offering at 14,319 units, each including 1,300 Series 2 options at NIS 7.2 per option, plus Series 3 and Series 4 options for no additional consideration. Expected immediate gross proceeds are about NIS 134 million, before about NIS 2.96 million of offering expenses.
Those proceeds are intended to finance the acquisition of 100% of SAGE, but they are only the first layer of the new capital structure. Full exercise of Series 3 options could add about NIS 75.2 million by October 1, and full exercise of Series 4 options could add about NIS 214.8 million by October 1, 2029. Series 2 options can be exercised until July 1 at an exercise price of NIS 0.3 per share. The future exercise proceeds are not cash that is already in the company.
The cost for shareholders is material dilution. Before the offering, the company had 16.13 million shares, and 49.52 million shares on a fully diluted pre-offering basis. After the offering and full exercise of all options, fully diluted shares reach 99.64 million, and the offered securities represent about 50.3% of issued and paid-up capital after full exercise of existing options. On May 31, the Tel Aviv Stock Exchange approved registration of the options and shares issuable upon exercise, subject to conditions, and a separate approval set a 60-day validity period for the options registration. Even after the offering is completed, the company does not automatically move back to the main list and will need to meet new-company rules if it wants to exit shell-company maintenance-list status.
Land and Power Still Need to Become a Buyer or Customer
The item that prevents the deal from becoming a celebration of gigawatts is the asset structure. SAGE mainly holds options to purchase land. Exercise of the full land-purchase rights across the projects is expected to cost about $137 million, on top of aggregate option fees of about $7.88 million through 2029. The company estimates first-phase development costs across the five projects at about $24-28 million, but these amounts exclude the land purchase price and the final deposit for the power-supply agreement if those are required before a sale.
The economic route is therefore not necessarily full data-center construction. The main described path is advancing the land and power package to Powered Land and selling it to a buyer that continues construction. Under that assumption, the company estimates that consideration for the first phase across the portfolio could reach $310-620 million in 2027-2029, based on a range of $200-400 thousand per MW IT. That is a large number, but it rests on estimates, comparable transactions and the assumption that a buyer is willing to pay for the rights once the site reaches the right stage.
The power plants make the gap between rights and required capital even clearer. If SAGE builds the relevant co-located plants itself, construction-cost estimates are about $1.68-1.82 billion for Sagebrush 2, about $600-650 million for Sagebrush 4 and about $1.44-1.56 billion for Sagebrush 5. These are not amounts financed by the current offering. They explain why the company points to a future mix of operating cash flow, equity and debt raises, and project finance, meaning debt supported by the project itself, depending on each project's progress.
Closing, Options and One Advancing Project Will Decide the Next Read
The Stark Power move deserves analysis because it adds a purchase agreement, a detailed portfolio and a financing structure, not only a hot sector label. The current read is that the company received a more real development asset than it had before the agreement, but it still has not shown a funded route to customers, sale proceeds or cash flow. The next disclosure that changes the read will be completion of the SAGE transaction and receipt of the offering proceeds. After that, option exercises, signed land rights and progress on grid connection or power agreements become the relevant proof points. Without one of those steps, the SAGE acquisition remains mainly an expensive option on US infrastructure, not an activity that returns cash to an Israeli public company.
Disclosure: Deep TASE analyses are general informational, research, and commentary content only. They do not constitute investment advice, investment marketing, a recommendation, or an offer to buy, sell, or hold any security, and are not tailored to any reader's personal circumstances.
The author, site owner, or related parties may hold, buy, sell, or otherwise trade securities or financial instruments related to the companies discussed, before or after publication, without prior notice and without any obligation to update the analysis. Publication of an analysis should not be read as a statement that any position does or does not exist.
The analysis may contain errors, omissions, or information that changes after publication. Readers should review official filings and primary sources before making decisions.