Zoho founder Sridhar Vembu has raised concerns about the sharp increase in computer memory prices and the growing cost of artificial intelligence infrastructure. According to Vembu, the combination of higher memory prices and rising AI related costs is making it increasingly difficult for technology companies to operate without reconsidering their pricing strategies.
Vembu recently highlighted a report indicating that memory prices had increased by around 500 percent over the previous 12 months. He also pointed to the growing cost of AI tokens as another factor putting pressure on technology businesses. He said Zoho had so far tried to avoid passing the additional costs on to customers, but maintaining that approach was becoming increasingly difficult.
Memory prices become a major concern
Computer memory is an essential component of smartphones, laptops, servers and data centre infrastructure. The rapid expansion of artificial intelligence has significantly increased demand for computing hardware, including memory.
Reports cited in recent coverage indicate that some DDR5 memory products have experienced dramatic price increases. Business Today reported that memory prices had risen by as much as 500 percent in a year, prompting Vembu to argue that the era of treating memory as an inexpensive resource may be coming to an end.
The increase has implications beyond technology companies. Higher component costs can eventually affect the prices of consumer electronics and enterprise hardware.
AI boom puts pressure on hardware
The rapid development of generative AI has resulted in major investments in data centres and computing infrastructure.
AI systems require large amounts of computing power and memory, particularly during the training and operation of large models. As technology companies expand their AI infrastructure, demand for processors, memory, storage, power systems and cooling equipment has increased.
Vembu has argued that the AI investment boom is putting pressure on several parts of the technology supply chain. He has specifically mentioned memory, CPUs, GPUs, electricity infrastructure, transformers, backup generators and cooling systems.
Impact on technology businesses
For software companies, higher infrastructure costs can create a difficult business environment.
Software businesses typically depend on servers, cloud computing, storage and other technology infrastructure. When the underlying cost of these resources rises sharply, companies must decide whether to absorb the additional expense or increase prices for customers.
Vembu said Zoho has attempted to hold prices steady despite rising costs. However, he acknowledged that maintaining existing prices is becoming harder as memory and AI related expenses continue to increase.
Possible impact on smartphones and laptops
The memory shortage could also affect consumers.
Smartphones and laptops require memory components, and manufacturers could face higher production costs when component prices rise.
Vembu has connected the rising prices of some consumer electronics with the broader AI investment boom. He argued that the huge spending on AI infrastructure is creating pressure throughout the technology supply chain.
If higher component costs continue for an extended period, manufacturers may eventually have to adjust product prices, reduce specifications or absorb lower profit margins.
The changing economics of AI
The AI boom has created enormous opportunities for technology companies, but it has also introduced significant costs.
Companies developing AI systems need expensive computing infrastructure, specialised chips, large quantities of memory and substantial electricity.
As AI services become more widely used, the cost of running those systems becomes an important business consideration.
Vembu's comments highlight the possibility that the economics of AI could become more complicated as companies move from experimentation to large scale commercial deployment.
Rising AI token costs
Memory is not the only concern raised by Vembu.
He has also pointed to rising AI token costs. AI tokens are units used to measure the amount of text processed by many AI models.
For businesses using AI extensively, higher token costs can increase the expense of operating AI powered applications.
Vembu said the combination of memory prices and AI token prices was making business increasingly difficult.
This could become an important consideration for companies that are adding AI features to software products while attempting to keep subscription prices competitive.
Efficiency may become more important
The rising cost of computing resources could encourage software developers to focus more heavily on efficiency.
For years, developers have often benefited from increasingly affordable and powerful computing hardware.
When memory and processing resources were relatively inexpensive, software could sometimes use more resources without creating a major increase in operating costs.
That environment could now be changing.
Vembu has argued that software developers may need to pay greater attention to memory efficiency as hardware becomes more expensive.
AI investment and the technology market
Vembu has also expressed a broader view about the current AI investment cycle.
He does not argue that artificial intelligence itself will fail. Instead, he has distinguished between the long term potential of AI technology and the investment boom surrounding it.
He has compared the current AI investment cycle with previous technology investment bubbles, including the telecom boom of the late 1990s.
His argument is that a technology can ultimately become successful even if companies and investors involved in an investment boom experience losses.
In other words, the usefulness of AI does not necessarily guarantee that every AI related investment will be profitable.
The broader supply chain impact
The AI boom is affecting more than memory manufacturers.
Large data centres require significant quantities of electricity and cooling equipment. They also require networking hardware, processors, storage systems and power management infrastructure.
As companies build more data centres, demand for these components can increase simultaneously.
This creates a potential supply chain challenge in which multiple industries compete for limited manufacturing capacity.
Vembu has described this as one of the broader effects of the current AI investment cycle.
What could happen to consumers
Consumers could eventually feel the effects of higher technology infrastructure costs.
If manufacturers face higher memory and component prices, they may respond by increasing product prices.
Another possibility is that manufacturers could maintain prices while offering lower memory configurations or other reduced specifications.
The actual impact will depend on how long the supply pressure continues and how manufacturers respond.
Software companies face a difficult balance
Technology companies must balance three major factors: rising infrastructure costs, customer expectations and competition.
Increasing subscription prices can help companies cover higher costs, but it can also make products less attractive to customers.
Absorbing the costs can protect customers but may reduce profit margins.
Companies may therefore increasingly focus on making their software and AI systems more efficient.
AI can still deliver long term benefits
Despite his concerns about the current investment boom, Vembu remains positive about the potential of artificial intelligence.
His argument is essentially that the technology can be useful and become widely adopted even if the current investment cycle eventually slows.
This distinction is important because rising costs do not necessarily mean that AI development will stop.
Instead, businesses may become more selective about where they invest in AI and how efficiently they use computing resources.
Importance for Indian technology companies
The issue is particularly relevant to India's technology sector.
Indian software companies increasingly use cloud infrastructure, AI services and large scale computing systems.
Higher global hardware costs could therefore affect operating expenses for companies serving customers in India and overseas.
At the same time, India's growing AI ecosystem is creating opportunities for startups and established technology companies.
The challenge will be to build AI products that provide enough value to justify their computing and infrastructure costs.
What companies may need to change
If memory and AI infrastructure costs remain elevated, companies may have to rethink how their products are designed.
Developers could place greater emphasis on efficient code, memory management, model optimisation and reduced computing requirements.
AI models could also be optimised to deliver similar results while consuming fewer resources.
Such improvements could become an important competitive advantage as infrastructure costs increase.
The end of cheap memory
Vembu's comments point toward a broader change in the technology industry.
For many years, declining hardware prices helped companies deliver increasingly powerful products at relatively affordable costs.
The current AI driven demand surge is putting pressure on that model.
If memory remains expensive, software and hardware companies may have to treat computing resources as a more valuable commodity.
This could influence product design, pricing and investment decisions across the technology sector.
Outlook
The rapid expansion of artificial intelligence is creating both opportunities and challenges for the technology industry.
AI companies are investing heavily in infrastructure, while demand for memory and other computing components continues to increase.
Sridhar Vembu's warning highlights the business pressure created by this trend. He says Zoho has tried to avoid increasing prices, but rising memory and AI related costs are making that increasingly difficult.
The reported 500 percent rise in memory prices over the past year is particularly significant, although the exact increase varies depending on the memory product and period being compared.
For businesses, the development could encourage greater emphasis on efficiency and cost control.
For consumers, prolonged increases in component costs could eventually contribute to higher prices for smartphones, laptops and other electronic products.
The AI boom therefore has an increasingly complex impact on the technology economy. While artificial intelligence could deliver major long term benefits, the cost of building and operating the infrastructure required to support it remains a growing challenge.