OKLO may be priced like a nuclear company today, but what it's actually building is AI-enabled infrastructure for the energy age. While the market is still preoccupied with short-term questions like whether Aurora reactors will be licensed and whether the first deployment will take place in 2027-2028, the real issue is that electricity demand is structurally exploding, and the current grid architecture cannot meet it. For AI data centers, high-performance computing (HPC), defense facilities, and critical infrastructure, the problem is no longer the price of electricity; it's its continuity, security, and scalability. This is where OKLO's difference comes in. Rather than being a producer selling energy to the classic grid, the company aims to be an infrastructure provider offering 24/7 clean energy guarantees directly to the customer, especially with onsite micro-reactors for hyperscalers and colocation data centers. OKLO's Aurora powerhouse technology, unlike large power plants, is built on a distributed energy architecture, not a centralized one. Thanks to its compact design with a capacity of 15-75 MW, fast neutron reactors that do not require refueling for many years (10+ years), passive safety features, and low operating requirements; OKLO offers on-site energy solutions for data centers, military bases, remote industrial facilities, and AI campuses. Its liquid metal-cooled, metal-fueled system differs from traditional water-cooled reactors. Inherent safety ensures automatic shutdown in case of accidents. This represents a paradigm shift, transforming nuclear power from a public utility into a private sector infrastructure solution. Collaboration with Vertiv dramatically increases the energy efficiency of data centers by integrating the reactor steam directly into cooling systems, creating an ideal closed-loop system for AI payloads.

OKLO isn't just designing reactors; it's also redefining the fuel cycle. The approach of producing fuel by recycling spent nuclear fuel and plutonium offers both cost advantages and a strong political narrative. Plutonium criticality experiments with Los Alamos and the fuel mill (A3F) approved by the DOE at Idaho National Lab are making this cycle a reality. If this model is scalable, which I believe it is—and the DOE pilot programs and rapid NSDA approvals in 2025 indicate this—nuclear energy will for the first time be in an advantageous position, not on the defensive, in discussions about nuclear waste issues. The pipeline has reached approximately 14 GW, largely coming from data centers. Landmark deals like the 12 GW agreement with Switch demonstrate the concretization of hyperscaler demand.

Looking at OKLO's balance sheet today, it's easy to say that revenue is zero and losses are high, because that's what I'm saying too. There's zero revenue and significant net losses in the first nine months of 2025, but the cash position is around $1.2 billion, and with the $1.5 billion ATM program, financing is strong, and share dilution will continue. However, this company needs to be evaluated not by today's figures, but by the energy nodes it can control in the future. Just as AWS became not just a business unit of Amazon, but the profit engine of the entire ecosystem; OKLO could become an indispensable energy layer for critical loads in the AI ​​age. The multiples here aren't classic utility multiples. The combination of a build-own-operate model + fuel recycling + regulatory barriers + data center integration creates a very different valuation set. I think the market value will be at a different level than it is today, with initial revenues in 2027-2028 (estimated starting at $10-20 million) and GWs of deployments by 2030.

The regulatory side seems daunting for many investors, but in the long run, this could work in OKLO's favor, not against it. The reason is the US energy crisis I discussed last week. NRC processes are difficult; yes, the previous application was rejected, but Phase 1 readiness was completed in 2025, the PDC topical report was accepted, and draft evaluation is expected in early 2026 with an accelerated timeline. DOE's Reactor Pilot and Fuel Line programs, groundbreaking in INL, and NSDA approvals are opening up alternative avenues. The ADVANCE Act and executive orders are also increasing support. An approved and operational fast reactor design can scale for years without facing serious competition, while competitors like $SMR NuScale struggle with cost and licensing issues (unlike the approved design). This reinforces the first-come, first-served dynamic, especially for grid-independent power solutions for AI data centers.

The story of OKLO and its potential return lies precisely here. OKLO is a company whose potential for failure is currently priced in (dilution, execution delays, regulatory risk), but whose potential for success is not yet fully written. If the energy bottleneck for AI data centers becomes more visible in the second half of the 2020s—and Nvidia CEO Jensen Huang's support for nuclear power and the hundreds of GW of demand expected by 2030 indicate this—solutions like OKLO will cease to be optional and become mandatory infrastructure. Switch, Vertiv, and undisclosed hyperscaler agreements (750 MW+) could be harbingers of this transformation.

Neither the US administration nor I view OKLO as a nuclear power company. OKLO should be viewed as a long-term call option on a structural break in electricity demand. It's a bet on the bottleneck of the AI ​​economy in the physical world: clean, reliable power. A regulated infrastructure story with very few competitors if scaled up: fast reactors + fuel recycling. It's not currently in my portfolio; its valuation is at a premium, but if the scenarios I described above materialize, OKLO's return would be disproportionate. The market is currently preoccupied with dilution fears and short-term volatility. OKLO, however, is moving in a completely different timeline: NRC/DOE milestones in 2026, initial operations in 2027-2028, and then GWs of deployments. This suggests that OKLO should be looked at for 2028, and small purchases should be made with DCA during a sharp correction.

With technological advancements (plutonium experiments, fuel plant) and government support, OKLO's role in the AI ​​energy revolution is steadily increasing. Yes, OKLO is a very good company, but I'll keep saying it's very expensive. OKLO is a tremendous company, but equally expensive. Being expensive doesn't mean the stock won't rise, but at these prices, it doesn't have a place in my portfolio, although as you know, it's on my watchlist.

The information, comments and recommendations contained herein are not within the scope of investment consultancy. Investment consultancy services are provided within the framework of the investment consultancy agreement to be signed between brokerage firms, portfolio management companies, banks that do not accept deposits and customers. The comments in this article are only my personal comments and these comments may not be appropriate for your financial situation and risk return. For this reason, investments should not be made based on the information and comments in my articles.