OpenAI chief executive Sam Altman has sparked criticism after suggesting that consumers holding large amounts of computer memory in their personal devices is an inefficient use of limited global resources. His comments come as the technology industry faces a growing RAM shortage, driven largely by increasing demand from artificial intelligence (AI) data centres.
During an appearance on the Cheeky Pint podcast, Altman argued that the limited supply of dynamic random-access memory (DRAM) would be better utilised in cloud infrastructure than in individual computers. His remarks have drawn criticism from PC enthusiasts and consumers already dealing with rising hardware prices and concerns about the wider impact of AI expansion.
The global memory shortage has become an increasingly serious issue as technology companies compete for components needed to build and operate AI systems. Data centres require substantial quantities of high-performance memory to support the computing workloads behind services such as ChatGPT. This growing demand has placed additional pressure on manufacturers and reduced the supply available to consumer electronics markets.
Altman questions how consumers use computer memory
Speaking on the Cheeky Pint podcast, Altman explained that the current distribution of memory resources was not ideal globally. He suggested that large quantities of DRAM installed in personal computers were not being used efficiently, particularly when compared with the demands of cloud-based computing infrastructure.
“From a global allocation perspective, actually, it’s quite terrible to have everybody have a bunch of the world’s limited supply of DRAM sitting in their own computer.”
Altman argued that individual computers often have more memory than their owners regularly need. In contrast, data centres supporting AI services can use available resources more consistently. He suggested that shifting more memory towards cloud infrastructure could improve overall utilisation during a period when supplies remain restricted.
“You want maximum utilisation in the cloud if you’re trying to deal with this very supply-constrained environment,” the OpenAI CEO continued. “And somehow many people’s intuition is the opposite of that.”
His comments reflect the growing importance of memory resources in the AI industry. Modern AI models require substantial computing capacity, with memory playing a central role in processing large datasets and responding to user requests. As companies expand their AI infrastructure, demand for specialised memory products has increased significantly.
However, the suggestion that consumers should have less access to computer memory is likely to be controversial. RAM remains an essential component for gaming, content creation, software development and everyday computing. Many PC users also purchase additional memory to improve performance, support demanding applications and extend the useful life of their machines.
The discussion also highlights a growing divide between AI companies’ priorities and those of ordinary technology consumers. While businesses are focused on securing enough hardware to expand their services, individual buyers are increasingly concerned about affordability and the availability of essential PC components.
PC enthusiasts push back against Altman’s comments
Altman’s remarks have received a largely negative response from online communities, particularly among PC enthusiasts and people concerned about the effects of AI expansion. A discussion on Reddit attracted criticism of the suggestion that consumers should give up their computer memory to support data centre operations.
One commenter said they wouldn’t surrender their existing hardware, writing: “It’s only 64GB, but he can pry it from my cold, dead hands”.
Another criticised the broader approach taken by major technology companies, stating that “[tech company executives] will destroy the planet by continuing with this mentality and then they’ll promptly blame everyone else for the downfall”.
Other responses questioned whether AI companies were putting their own infrastructure needs ahead of consumers. The discussion also raised concerns about the environmental consequences of expanding data centres, including their electricity and water consumption, as well as the potential pollution associated with their operations.
Concerns about the ethics of AI development also featured in the debate. Some users pointed to the use of copyrighted material in training datasets, arguing that technology companies have benefited from large amounts of existing creative and intellectual work without always securing permission or compensating the original creators.
These criticisms reflect wider frustration with the rapid growth of AI services. Although AI tools have introduced new capabilities across several industries, their expansion has also created challenges involving infrastructure costs, employment, energy consumption and access to computing hardware.
For PC owners, the prospect of competing with large technology companies for limited memory supplies is particularly concerning. Data centre operators can purchase hardware in enormous quantities, potentially giving them greater influence over manufacturing priorities and supply agreements than individual consumers or smaller computer retailers.
Altman’s comments have therefore added another point of tension to an already difficult situation. While his argument focuses on improving the efficiency of global memory resources, consumers may see the proposal as another example of AI companies expecting ordinary users to accommodate the industry’s growing demands.
Memory shortage could continue affecting consumers until 2030
The global RAM shortage is not expected to disappear quickly. Several industry analysts have warned that supply constraints could continue for years as memory manufacturers struggle to keep up with demand from AI infrastructure projects and other technology markets.
Forecasts suggest the shortage could extend toward 2030, creating prolonged uncertainty for PC manufacturers, hardware retailers, and consumers. Although additional production capacity may eventually improve availability, expanding semiconductor manufacturing requires significant investment and time.
The situation is particularly challenging because memory manufacturers must balance demand from different customer groups. AI data centres increasingly require high-bandwidth memory and other advanced components, while traditional consumer products continue to depend on standard DRAM. As manufacturers prioritise more profitable or strategically important markets, the supply available for personal computers could remain under pressure.
The consequences are already becoming a concern for people planning to build or upgrade their computers. Higher memory prices can increase the overall cost of a PC, particularly when combined with rising prices for other components. Gaming systems, workstations and computers used for professional applications may become more expensive as manufacturers and retailers adjust their prices to reflect supply conditions.
Smaller computer builders could also face additional difficulties competing with major companies that have greater purchasing power. Consumers may have fewer affordable options when selecting memory configurations, potentially forcing them to delay upgrades or settle for lower specifications.
The shortage also raises questions about how the technology industry will manage the competing demands of AI development and traditional computing. Data centres are becoming increasingly important to the global digital economy, but personal computers remain essential tools for millions.
While Altman’s argument centres on maximising the use of limited memory resources, consumers suggest that many are unwilling to accept reduced access to hardware to support the expansion of AI services. With supply constraints expected to persist, the debate over who gets access to the world’s available memory is likely to continue.
