Samer Choucair: Rising Memory Costs Are Reshaping the AI Investment Landscape
Investment leader Samer Choucair said Nvidia’s move toward raising prices on some artificial intelligence servers by more than 15% is an important signal that the AI investment cycle is entering a new phase, one in which suppliers of critical inputs have greater power to pass rising costs on to customers.
Samer Choucair explained that the increase, linked to higher costs for high-bandwidth memory chips and expected to appear in systems shipping from early 2027, reflects more than temporary inflationary pressure. In his view, it points to a shift in bargaining power across the AI infrastructure supply chain.
“The speed at which these costs are being passed through suggests that the AI investment cycle has entered an early stage of maturity, where margins can no longer absorb every inflationary pressure across the supply chain,” Samer Choucair said.
Choucair added that rising demand for HBM memory, combined with constrained production capacity among major manufacturers, is increasing the cost of advanced servers and creating new capital-expenditure challenges for data-center operators and cloud-computing companies.
Reassessing AI Investment Models
Choucair said higher server prices could force large technology companies to reconsider the timing and sequencing of data-center spending, particularly if financing costs remain elevated.
He added that the impact extends well beyond Nvidia and memory manufacturers. Data-center operators, cloud-service providers, semiconductor companies, and developers of power and cooling infrastructure are all exposed to the changing economics of AI infrastructure.
According to Choucair, memory producers could benefit from stronger pricing power, while cloud-computing providers may face additional pressure on margins. That dynamic could encourage hyperscalers and other technology companies to seek more efficient system architectures or develop alternative chips designed to reduce dependence on expensive components.
“Institutional investors are increasingly looking for AI exposure through parts of the value chain that are less sensitive to input-cost inflation, including software and edge solutions, rather than relying exclusively on expensive physical infrastructure,” Choucair said.
New Opportunities in Semiconductors
Samer Choucair said higher memory costs could create investment opportunities for companies developing model-compression technologies, more efficient computing architectures, and solutions that reduce memory and power requirements.
Companies with vertical integration or long-term supply agreements could also become more attractive, particularly if they can pass higher input costs on to customers without materially weakening demand.
Choucair said these trends could support greater private-equity and venture-capital activity in semiconductors, specialized computing, energy efficiency, and advanced cooling technologies.
The underlying investment question, he argued, is shifting from simply identifying who benefits from rising AI demand to determining which companies can preserve margins while infrastructure costs increase.
Implications for Saudi Arabia and the Gulf
Samer Choucair said the rising cost of AI infrastructure is particularly relevant to Gulf economies, led by Saudi Arabia, as investment in data centers and the digital economy expands under Vision 2030.
Higher server prices increase the cost of building domestic digital capacity, but they simultaneously strengthen the case for diversifying technology suppliers, negotiating longer-term procurement partnerships, and investing in more efficient infrastructure.
“Countries building digital infrastructure today have an opportunity to renegotiate long-term supply arrangements or invest in alternatives that reduce exposure to volatility in memory prices,” Choucair said. “That is a fundamental part of risk management in strategic capital allocation.”
He added that Saudi Arabia could benefit by allocating a greater share of investment toward the infrastructure surrounding AI rather than concentrating only on the purchase of computing capacity.
That broader ecosystem includes power generation, cooling systems, data centers, networking infrastructure, and advanced software. As AI workloads become increasingly capital intensive, the economic value of these supporting layers is likely to grow.
Risks and Opportunities
Choucair cautioned that sustained increases in equipment prices could delay some data-center projects or force companies to reprioritize capital spending, particularly where projects rely heavily on debt financing.
At the same time, he said the structural strength of demand for artificial intelligence makes a collapse in the investment cycle unlikely. Instead, the composition of investment could shift toward technologies that deliver greater efficiency and are less exposed to supply-chain bottlenecks.
“The real opportunity lies in distinguishing cyclical growth from structural growth,” Choucair said. “The current increase in costs confirms that underlying demand remains strong, but it also imposes greater discipline when evaluating risk-adjusted returns on AI infrastructure investment.”
That distinction is becoming increasingly important for institutional investors. Rising demand does not necessarily guarantee attractive returns at every point in the value chain, particularly if capital costs, component inflation, and competitive intensity rise faster than revenues.
A Broader AI Allocation Strategy
Samer Choucair concluded that the direction of the technology sector in 2027 will depend heavily on its ability to balance continued demand growth with inflation in the cost of critical components.
For institutional investors, that means AI exposure is likely to become more diversified across semiconductors, memory, energy, data centers, and software rather than being concentrated in a single layer of the value chain.
Choucair said this broader allocation approach could become increasingly important as artificial intelligence moves from an early expansion phase into a more capital-disciplined stage of development.
In that environment, the strongest investment opportunities may no longer be defined simply by which companies are growing fastest, but by which businesses can secure critical inputs, preserve margins, improve efficiency, and generate durable returns as the cost of building AI capacity continues to rise.