Artificial Intelligence

OpenAI CEO Sam Altman Criticizes PC RAM Allocation

In a recent podcast interview, OpenAI CEO Sam Altman argued that consumer computers having local DRAM is inefficient for global allocation, suggesting that memory resources are better utilized in cloud data centers.

Radwan AbdulhadiEditor and technology writer

Published: October 31, 2026 · Updated: October 31, 2026 · 3 min read

OpenAI CEO Sam Altman Criticizes PC RAM Allocation — Image: Tom's Hardware

OpenAI CEO Criticizes Home Computer Memory

OpenAI CEO Sam Altman recently addressed the topic of local inference and consumer hardware during an interview over beers with the YouTube channel Cheeky Pint, stating that the widespread presence of DRAM in home computers is 'quite terrible' from a resource perspective.

While discussing open-source models and the assumption that everyday users want to run local artificial intelligence on their own custom or pre-built machines, Altman explained why he views this setup as flawed.

Global Allocation and Cloud Utilization

Elaborating on his viewpoint, Altman emphasized that in a supply-constrained environment, global allocation suffers when individual users maintain a share of the world's limited DRAM supply inside their personal systems.

According to the OpenAI chief executive, home computer memory often sits idle for significant portions of the time, whereas cloud architectures strive for maximum utilization.

'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 and being used 5% of the time,' Altman stated during the interview. He added that the goal should be achieving maximum utilization in the cloud when confronting supply constraints.

Microsoft data center in Mount Pleasant, Wisconsin — (Image credit: Microsoft) · Source: Tom's Hardware

Business Alignment and Open-Source Dynamics

Although OpenAI features some open-source offerings, its primary business model relies heavily on closed-source models hosted within centralized data centers. Consequently, Altman's remarks align directly with the company's operational structure.

As local AI development advances, powerful home gaming rigs and laptops could theoretically enable users to run capable models locally, potentially reducing the need for costly commercial cloud subscriptions.

Altman's commentary implies that memory resources would generate greater value inside massive data center deployments rather than supporting conventional desktop applications like web browsing, spreadsheets, or gaming.

Why Sam Altman is against a single superintelligence - YouTube — Image related to the report from Tom's Hardware · Source: Tom's Hardware

Enthusiast Backlash and Hardware Ownership Concerns

The remarks have generated substantial discussion across hardware enthusiast communities, raising concerns about a potential shift away from local hardware ownership.

Critics and technology commentators noted that prioritizing cloud-exclusive infrastructure could push consumers toward renting virtual computing power and AI agents rather than buying and maintaining their own physical components.

Beyond artificial intelligence discussions, the wide-ranging interview also touched upon other technical topics, including mathematical research at OpenAI, model sandboxing techniques, and recent software security vulnerabilities.