The way Navitas Semiconductor has been trading lately has an almost theatrical quality. After spending the majority of the previous year hovering around the price of a sandwich, the stock closed Friday at $18.20, up more than 15% in a single session. It was at $1.83 a year ago. Because Navitas’s actual income statement doesn’t provide a clear explanation, that’s the kind of move that makes traders sit up, analysts hedge their language, and serious investors a little uneasy.
The company produces silicon carbide and gallium nitride power chips, which are the kind of parts that most people will never come into contact with, despite the fact that they are subtly becoming essential inside the racks of AI data centers. Now that a single Nvidia server rack can draw the power of a small neighborhood, GaN and SiC allow electricity to flow with less loss and less heat than traditional silicon. Anyhow, that’s the pitch. And since Navitas was selected as a power semiconductor partner for Nvidia’s next-generation AI factory platforms, the pitch has grown in volume.
The Q1 2026 figures, which were released earlier this month, present a more nuanced picture. At $8.6 million, revenue was down almost 39% year over year but up 18% sequentially. The segment management consistently highlights high-power markets, which have grown by about 35% annually and now account for the majority of sales. $27.8 million was the quarter’s GAAP operating loss. The new CFO, Tonya Stevens, made a point of bringing up the $221 million in cash and zero debt, which seems like the kind of information you bring up when people start asking awkward questions about runway.
CEO Chris Allexandre informed investors that the company is “meaningfully reaccelerating” its transition away from low-end consumer electronics and mobile chargers. He did not say that the pivot was complete. “Too early to declare victory,” he said, which is probably the most honest line out of any chip executive this quarter. The company continues to discuss a $450 million pipeline, but it’s still unclear if design will prevail and whether the product will be shipped on time enough to support a $4.25 billion market capitalization.
The analyst environment surrounding the stock is peculiar. Following the earnings, Needham increased its price target to $21. However, some companies have price targets in the single digits, which were set months ago when Navitas was a completely different story, and the broader consensus target is significantly lower than the current price. In other words, Wall Street hasn’t caught up to the decisions made weeks ago by a few Reddit boards, momentum funds, and retail traders. No one was able to calibrate during the short squeeze in April, when shares surged 88%.
It feels like a familiar cultural moment around Navitas. The early days of Plug Power, FuelCell Energy, and a dozen other small businesses that unexpectedly found themselves enmeshed in a much larger story—sometimes justifiably, sometimes not—are echoed. In the power semiconductor industry, Infineon, Texas Instruments, Onsemi, and STMicroelectronics all have much larger budgets and more established clientele. Integration and timing are Navitas’s advantages, and in this industry, timing is the most delicate factor. According to management, Nvidia’s Blackwell rack systems will significantly increase in revenue in 2027. Asking shareholders to wait that long is excessive, particularly at these prices.
It’s hard not to notice the gap between the technology demos and the income statement. The 800-volt power boards, the solid-state transformer prototypes, the SiC MOSFET launches — they’re genuinely impressive engineering, the kind of thing you’d want a real chip company to be doing. There is currently no answer to the question of whether that engineering turns into a real business on a real timeline. Perhaps that’s the most truthful thing anyone can say about Navitas stock at the moment.
