Zoho co-founder Sridhar Vembu has warned that memory component prices have spiked 500% over the past year, creating significant operational strain. The surge, driven by intense demand for AI infrastructure, is testing the sustainability of existing pricing models. This cost pressure is forcing technology companies to re-evaluate software design to rely less on cheap, abundant memory.
Zoho co-founder Sridhar Vembu has highlighted a significant challenge facing the technology sector, pointing to a 500% surge in memory component prices over the last 12 months. This sharp increase in hardware costs, which Vembu described as rising tenfold from previous lows, is placing mounting pressure on operational budgets for software and technology firms globally.
The price hikes are largely attributed to the massive, ongoing demand for Artificial Intelligence (AI) infrastructure. Chip manufacturers are increasingly shifting their production focus toward high-bandwidth memory (HBM) and other specialized components required for AI data centers. This strategic shift has created supply shortages for standard consumer and enterprise memory components, driving up prices across the board.
Evidence of this market strain is visible in the hardware sector, where standard DDR5 memory kits—a common component in servers and high-performance computers—have seen year-on-year price increases ranging from 350% to nearly 500%. For businesses, these procurement costs are no longer negligible and are beginning to affect the financial feasibility of maintaining existing pricing models for digital products and services.
Beyond immediate cost concerns, the situation is prompting a shift in how software is developed. For decades, the industry has operated under the assumption that memory capacity would remain cheap and abundant, allowing software to be built without stringent memory constraints. Vembu has argued that this era has ended, necessitating a return to memory-efficient programming languages and smarter system compilers. The focus, he suggests, must shift toward developing software that delivers high performance and safety without demanding excessive hardware resources.
For the broader technology sector, this represents a structural change. Companies that have traditionally relied on low-cost hardware to scale their services may face margin pressure as component costs stay high. The sustainability of the current AI infrastructure investment boom remains a key point of discussion. If the cost of building and operating AI systems continues to rise at this pace, businesses may eventually have to pass these costs on to end-users or find ways to significantly optimize their hardware usage.
Investors and industry watchers will be monitoring whether these hardware prices stabilize or continue to climb as AI infrastructure spending evolves. The ability of companies to optimize software efficiency and manage these rising infrastructure costs will be a critical factor in maintaining operational health in the coming quarters.