On June 1, 2026, shares of global grid-scale energy storage leader Fluence Energy (FLNC.US) soared 43.8% to close at $27.15, marking the highest closing price since February. The rally was triggered by a single event: Siemens officially released its reference electrical and power architecture for AI factories built on the NVIDIA DSX Vera Rubin NVL72 platform, naming Fluence’s SmartStack battery energy storage system as the exclusive battery storage partner in the design.
This marks the first time energy storage has been explicitly written into the top-level blueprint of an AI data center. It signals that energy storage is no longer just a "supporting component" for renewable energy projects—it’s becoming a "standard part" of multi-billion-dollar AI infrastructure.
The Macro Backdrop: Why Energy Storage Is Now Essential for AI Data Centers
To understand the rationality behind Fluence’s surge, we first need to examine the structural roots of the AI data center energy crisis.
According to the International Energy Agency (IEA), global data center electricity consumption will reach 415 TWh in 2024, accounting for 1.5% of worldwide electricity use. By 2030, this figure is expected to double to 945 TWh—comparable to Japan’s annual power consumption. Even more significant is the growth differential: between 2024 and 2030, US data centers will account for nearly 50% of the country’s incremental power demand. By 2030, US AI data processing will consume more electricity than the combined usage of traditional energy-intensive industries such as aluminum, steel, cement, and chemicals.
Morgan Stanley projects a cumulative US data center power shortfall of 47 GW from 2025 to 2028. In other words, insufficient power supply has become the primary bottleneck restricting AI computing expansion.
The role of battery energy storage in this context is undergoing a fundamental shift. Previously, energy storage’s value in data centers was mainly as a backup power source (replacing or supplementing UPS systems). However, the exponential increase in AI rack power density has changed the equation. Over the six years from NVIDIA’s A100 to Vera Rubin, single-GPU power consumption has jumped from 400W to over 1,400W, and single-rack power has climbed from 30kW to nearly 180kW. This surge brings two major challenges: First, the traditional six-stage UPS architecture (with only about 89% end-to-end efficiency) struggles to support such high-density DC loads. Second, GPU power spikes can reach 1.5 to 2 times average consumption, and the grid cannot smooth out these millisecond-level fluctuations using conventional methods.
As a result, energy storage systems in AI data centers have shifted from "backup" to "active support"—delivering grid-level functions like black start and voltage/frequency ride-through, while also smoothing AI load fluctuations. This is the core logic behind Fluence SmartStack’s inclusion in the reference design.
Fluence Energy (FLNC): The Real Value of Being the "Sole Designated" Partner for NVIDIA and Siemens
Breaking Down the Partnership
The Siemens DSX Vera Rubin AI factory reference design covers a total facility capacity of 136 MW, with 100 MW allocated to IT loads. The power chain runs from 34.5 kV municipal high-voltage access all the way to server racks. Fluence’s SmartStack energy storage system is configured at 120 MW/240 MWh, providing key functions such as voltage/frequency ride-through, black start, grid demand response, and AI load smoothing.
It’s important to note that this reference design is not a "contract," but a non-binding technical blueprint jointly released by Siemens and NVIDIA. However, Fluence is the only "designated BESS partner" in the blueprint, and among NVIDIA’s seven foundational infrastructure OEM designs, only Fluence is explicitly named.
The structural reason for this exclusivity is straightforward: Fluence is a joint venture between Siemens and US utility AES Corporation. Inclusion in the parent company’s official reference design is essentially a result of resource alignment, not a "technical competition" victory but an "ecological inheritance."
Why Did the Market Award a 43.8% Surge?
Beyond the endorsement effect of being the "sole designated" partner, the market also focused on a key parameter: the reference design specifies a battery runtime of 2–3 hours, much higher than the previous market consensus of around 1 hour. This means higher BESS demand and larger contract values per project. In addition, Fluence has already signed master service agreements with two hyperscale data center operators, bringing its order backlog to a record $10.1 billion—providing substantial visibility for future revenue.
Financials: Revenue Growth vs. Persistent Losses
In fiscal Q2 2026, Fluence posted revenue of $464.89 million, up 7.71% year-over-year, but well below the consensus estimate of $614.93 million. Net income was a loss of $20.927 million. Over the trailing twelve months, total revenue was about $2.58 billion, but TTM net income remained negative, with a P/E ratio of -59.66. The debt-to-equity ratio stands at 87.73%, and EBITDA is still negative.
In short, Fluence is a fast-growing but still unprofitable company. The 43.8% single-day surge was not based on near-term profitability improvement, but rather on upgraded expectations for future order conversion.
Is the Market Overreacting?
This is the article’s central analytical question. Let’s break it down:
Supporting a "reasonable" reaction: Being included in NVIDIA’s reference design is tantamount to a joint endorsement from NVIDIA and Siemens. This blueprint will likely become the go-to template for hyperscale data center builders. If each 136 MW AI data center follows this standard and installs 120 MW/240 MWh of storage, the value per BESS project could range from tens to hundreds of millions of dollars based on current system costs. Given the potential scale of global AI data center construction, this "locked-in ecological niche" carries significant long-term value.
Pointing to "excess": The reference design is currently non-binding and has not yet translated into actual orders. While Fluence’s revenue growth is accelerating, its gross margin is only about 11.71%, and it remains to be seen if scale can drive positive net profits. As of early June, the 12-month average analyst price target was just $18.59—about 25% below the post-surge share price. The surge pushed the stock above some analysts’ target ranges—Canaccord Genuity’s "buy" rating target is $28, and the June 1 close of $27.15 is nearly there.
Fluence’s 43.8% jump is logically grounded in "conceptual reaction," but there’s clearly some "pricing ahead of fundamentals." The narrative of energy storage becoming part of AI data center top-level design has strong long-term effects, but the path from concept to broad adoption remains uncertain in the short term. From an investment perspective, Fluence’s core risks are not on the demand side (which is largely validated), but rather: 1) when profitability will turn positive, and 2) to what extent "sole designation" can be converted into exclusive market share.
Peer Comparison: Three Other "AI Power" Stocks from Different Angles
Bloom Energy (BE): Distributed Fuel Cells for AI Data Centers
Fluence’s story is about "grid-integrated storage." Bloom Energy’s story is entirely different: its solid oxide fuel cells (SOFC) are standalone, off-grid power units that can be deployed directly at data center campuses and delivered within 90 days.
In Q1 2026, Bloom Energy reported revenue of $751 million, up 130.37% year-over-year and far exceeding market expectations ($540 million expected). Adjusted EPS reached $0.44, up more than 400% year-over-year. Full-year revenue guidance was raised to $3.4–3.8 billion, about 80% growth. Exploding AI data center orders are the main driver—about one-third of product orders now come from AI data center clients, and long-term service contracts total $9.6 billion.
But Bloom Energy faces a risk that is the polar opposite of Fluence’s: extreme valuation distortion. The company’s P/E (TTM) is around 11,970x, and the price-to-sales (P/S) ratio is about 34x. According to Gurufocus fair value estimates, BE’s current price (around $285) is about 2,760% above fair value ($10.27). The issue isn’t fundamentals—the demand for distributed AI data center power is real and strong—but rather that the current valuation already prices in extremely high growth, leaving virtually no room for a disappointing quarter.
NextEra Energy (NEE): A Scaled Platform for Renewables + Storage
Unlike the previous two, NextEra Energy is a fully mature and profitable utility giant. It operates two main segments: regulated utility Florida Power & Light (FPL) and NextEra Energy Resources, its clean energy development arm.
In Q1 2026, NEE delivered $6.7 billion in revenue, up 21% year-over-year; net income was $2.18 billion, up 162%. The company added 4 GW of new renewable and storage orders in the quarter (1.3 GW from batteries), bringing its backlog to about 28–33 GW. NEE reaffirmed its 2026 adjusted EPS guidance of $3.92–$4.02.
NEE’s strengths are risk diversification and dividend yield (currently about 2.6%), but its earnings upside is relatively limited. As AI-driven power demand grows, the market has awarded NEE a valuation premium (forward P/E about 23.7x, above the two-year average of 21.4x), but regulated utility assets naturally have less upside than pure-play AI power infrastructure names.
Vistra Energy (VST): Traditional Generator Riding the AI Power Wave
Vistra is one of America’s largest competitive power producers, operating a diverse fleet of natural gas, nuclear, coal, solar, and battery storage assets. Its benefit is the most direct: AI data centers are driving up US wholesale power prices, and as a net seller, Vistra fully benefits from this widening spread.
In Q1 2026, Vistra posted $5.64 billion in revenue, up more than 43% year-over-year; net income was $1.03 billion; and adjusted EBITDA reached $1.494 billion. The company completed a $6.13 billion share buyback and received investment-grade upgrades from two major rating agencies. Vistra expects 2026 core adjusted EBITDA from continuing operations to be $6.8–7.6 billion, a major increase over 2025 guidance.
Vistra’s risks are high leverage and earnings volatility due to wholesale power price swings. However, based on Q1 2026 results, its performance delivery is the strongest among the four stocks.
Comparative Matrix
| Dimension | Fluence (FLNC) | Bloom Energy (BE) | NextEra (NEE) | Vistra (VST) |
|---|---|---|---|---|
| AI Data Center Exposure | Storage in NVIDIA reference design | Distributed SOFC, 90-day deployment | Large-scale renewables + storage orders | Direct beneficiary of wholesale price increases |
| Recent Revenue Growth | +7.7% (Q2 2026) | +130% (Q1 2026) | +21% (Q1 2026) | +43% (Q1 2026) |
| Profitability | Loss, negative TTM net income | P/E ~11,970x | P/E ~24.7x, profitable | Profitable, high cash flow |
| Core Risk | Not yet profitable + order conversion uncertainty | Extreme valuation bubble | Limited upside elasticity | Price volatility + high leverage |
| Key Catalyst | NVIDIA reference design adoption | AI data center distributed power orders | Backlog conversion of clean energy orders | Sustained rise in wholesale power prices |
In summary, these four stocks represent four distinct risk-reward profiles within the AI power sector:
- Fluence: The most "exciting" concept, with the clearest ecological niche, but the longest path to profitability.
- Bloom Energy: Most direct realization of AI data center power demand, but valuation already reflects very high expectations.
- NextEra: The most stable long-term allocation, though with limited return upside.
- Vistra: The clearest path for AI-driven demand, strongest Q1 performance, but requires monitoring due to high leverage and volatility.
Latest Update: Gate Launches US Stock Trading, Bridging Crypto and Traditional Markets
For energy storage investment opportunities to become actionable, investors need convenient trading access. In June 2026, Gate officially launched US stock trading, achieving a key breakthrough by allowing users to trade US stocks directly with USDT.
Gate Stocks’ core features include: users can trade over 10,000 leading US stocks and ETFs through a unified Gate account—no need to switch to a traditional broker or open a separate account, with orders executed directly on major US exchanges. Dividends are automatically paid in USDT, with no funding or overnight fees.
Notably, Gate has fully integrated stock trading into its platform-wide VIP tier system—users with $2,000 in holdings qualify for VIP status and enjoy an exclusive minimum stock trading fee of 0.023%. Trading hours have expanded from the standard 6.5×5 to 16×5, covering pre-market, regular, and after-hours sessions, allowing investors to react instantly to major news (such as pre-market events like the Fluence announcement). Fractional share trading from as little as 0.01 shares is also supported, significantly lowering barriers to high-priced tech stocks.
For the four stocks discussed in this article (FLNC, BE, NEE, VST), investors can trade directly with USDT via Gate Stocks, managing both crypto and traditional securities seamlessly within a single account.
Conclusion
The AI data center energy crisis is reshaping how markets value energy storage, power, and new energy infrastructure. The joint NVIDIA-Siemens reference design, which incorporates storage into the core AI factory architecture, signals a transition from "concept" to "standard configuration" for the sector. Fluence Energy’s 43.8% single-day surge made it the focal point of this paradigm shift, but its ongoing losses and the non-binding nature of the blueprint mean there is a degree of pricing ahead of fundamentals.
Looking across the sector, Bloom Energy is the most direct beneficiary of distributed AI data center power demand but carries the greatest valuation risk; NextEra Energy offers long-term value through a stable utility base and a large storage order backlog; Vistra Energy stands out for the strongest Q1 fundamentals. Together, these four stocks represent the full spectrum of "high expectation" to "high delivery" in the AI power sector.
With Gate’s launch of US stock trading, investors can now easily access these opportunities across markets. However, no matter how compelling the long-term energy storage narrative, short-term price swings are still subject to the complex interplay of order conversion, interest rates, and market sentiment. Any sharp price move triggered by a single event must always be viewed in the context of underlying fundamentals and valuation constraints.
