AI is rewriting the simulation chip cycle: this time it's not just "restocking"

After more than three years of downturn, the analog semiconductor sector is back in investors' sights. According to the latest research by Morgan Stanley, the fundamentals of analog chips are moving from an "L-shaped bottom" towards recovery: channel inventory is becoming leaner, pricing pressures are easing, and supply of mature processes and power-related products is selectively tight. More importantly, the construction of AI data centers is pushing analog chips from traditional stories of automotive and industrial cycles towards a new narrative of "power and connectivity content enhancement." This is the biggest difference between this round of the market and previous rounds of analog chip cycles. In the past, analog semiconductors relied more on the recovery of industrial, automotive, and consumer electronics demand, but this round of recovery is supplemented by the expansion of AI infrastructure. Morgan Stanley pointed out in the report that AI computing and data centers are bringing new opportunities for rack power, digital power, storage interfaces, optical interconnection, and other areas, with benefiting segments including 800V power architectures, power conversion, MCUs, silicon photonics, optical interconnects, and low-earth orbit satellite communications. Signals of recovery: Inventory, pricing, and demand are improving simultaneously Looking at the industry cycle, analog semiconductors have been under pressure for a long time due to inventory clearance, declining prices, and weak terminal demand. But Morgan Stanley believes that the current inventory environment has reached a turning point. Their distributor surveys show that customer and distributor inventories are leaner, some product lead times have been extended by 4 to 8 weeks, the percentage of respondents planning to continue reducing inventory has significantly decreased, and in the analog, MCU, and connector product lines, there is a willingness to replenish inventory. There is also marginal improvement on the pricing side. The Morgan Stanley report indicates that the proportion of distributors with stronger pricing power for analog chips has increased from 33% to 65%, and for MCUs, the proportion with stronger pricing power has increased from 33% to 58%. In addition, Analog Devices implemented a price increase in February 2026, and Renesas hinted at possible price adjustments if raw material and logistics costs continue to rise. Rising prices of mature process wafer foundry services further enhance the ability of analog/MCU manufacturers to pass on costs downstream. AI data centers bring analog chips "from behind the scenes to the forefront" In the past, discussions about AI hardware in the market mainly focused on GPUs, HBMs, and advanced packaging. However, as the scale of AI data centers expands, the importance of power management, signal chains, interface chips, and optical interconnects is rising. American stock investment websites also mentioned when analyzing Texas Instruments that AI data centers are elevating analog demand, especially power and signal chain demands, to a new level; compared to GPUs/HBMs, data center analog devices are more likely to show a "broader spectrum, more robust, and longer-term" recovery pace. Morgan Stanley is particularly interested in the opportunities brought by the evolution of AI rack power architecture to 800V. The report states that in the next 12 to 24 months, changes in AI rack power architecture will become one of the most important structural variables in the analog chip field. Taking the example of NVIDIA's related rack architecture, the power side cabinets integrate traditional power supplies, rack-level battery backup units, and high-capacity capacitors, driving demand for SiC, GaN, solid-state transformers, and other power semiconductors. The report predicts that the power semiconductor content in each Rubin Ultra rack could exceed $20,000. Company clues: Texas Instruments, Renesas, NXP, and others provide cross-validation This resurgence is not only supported by macro narratives but also receives validation at the company level. The Morgan Stanley report believes that Texas Instruments, as the most broadly covered analog leader, has provided clear signals in terms of demand, pricing, and the outlook for industrial and data center prosperity. 21 Finance also mentioned that Morgan Stanley raised its target price for Texas Instruments from $180 to $221, reflecting institutional recognition of the improvement in data center and industrial demand. Renesas Electronics exemplifies the growth opportunities in digital power and storage interfaces. The report states that the company's short-term demand is strong, and it plans to increase channel inventory while investing 94 billion yen in capital expenditures to expand digital power capacity for data center applications. NXP reinforces the recovery logic in automotive and MCUs, with its second-quarter revenue guidance showing a sequential growth of 8.5% and a year-on-year growth of 18%, and it has disclosed for the first time its data center-related revenue exposure. STMicroelectronics is benefiting from optical, low-earth orbit satellite communication, and margin recovery. Reports from Juheng.com and East Money Information show that Morgan Stanley has recently raised its target prices for several semiconductor companies, shifting the logic from simply being "driven by AI concepts" to "AI and traditional cyclical demand jointly driving profit recovery." Among them, companies such as RDL, GlobalFoundries, are also seen as beneficiaries of stable demand, improved capacity utilization, and opportunities in silicon photonics. Risks remain: Don't misinterpret early cyclical signs as a full reversal However, Morgan Stanley does not define this resurgence as a risk-free market situation. The report points out that the biggest bearish argument is that analog semiconductors may just be normal inventory replenishment after a deep adjustment, rather than a sustained recovery in end-demand. If distributors complete inventory replenishment but industrial orders weaken, automotive production declines, or worsening pricing due to intensified competition in China, valuation revaluation may be hindered. Moreover, AI data centers themselves also carry financing and execution risks. Futu NiuNiu has cited Morgan Stanley's view that AI capital expenditures are rapidly expanding, with increasing reliance on debt financing by some large tech companies; if credit markets tighten, the pace of AI infrastructure construction may slow down. Economic Times Enterprise AI reported that Morgan Stanley believes Agentic AI will drive chip spending from GPUs to CPUs, memory, and other more extensive hardware segments, which is consistent with the logic of analog chips benefiting, but also means that related demand still depends on whether AI applications can continue to grow. Overall, analog semiconductors are at a crossroads of "cyclical bottom recovery + AI content enhancement." This round of the market may not necessarily require a V-shaped rebound in industrial and automotive demand, as long as inventory normalizes, prices stabilize, capacity utilization improves, and visibility of revenue from AI power and interconnects increases, leading companies have room for an upward revision in profit centers and valuation multiples. What really needs to be tracked is whether inventory replenishment can end-demand sales, and whether the supply-demand tension in AI data center power, interconnects, and mature processes can continue to materialize.