Altman Calls Local Hardware Memory Use Inefficient

OpenAI's CEO advocates for shared memory systems to solve global DRAM waste as hardware prices surge up to seven times previous levels.

OpenAI CEO Sam Altman is sounding the alarm on global hardware inefficiency, arguing that the current model of localized computing is a waste of critical resources. Speaking in late October 2026, Altman characterized the standard practice of equipping individual computers with massive amounts of DRAM as "terrible" from a global allocation perspective, noting that most of this memory sits idle approximately 95% of the time.

While Altman clarified in a conversation with Semafor that he still wants people to have access to significant RAM and believes "home tinkering is great," he maintains that the industry must move toward a more efficient architecture. The optimal solution, according to Altman, is a transition to shared memory systems where hardware can operate at 100% utilization rather than being siloed in underused personal devices.

A Growing Squeeze on Silicon

This push for efficiency follows Altman's earlier predictions of a "huge squeeze" on memory markets. That forecast appears to have materialized, with the CEO noting that DRAM prices have surged by as much as five to seven times their previous levels. This volatility highlights the precarious nature of the hardware supply chain as the industry scales to meet the demands of the AI revolution.

Altman was quick to credit the foundational role of hardware partners in this era, stating it is "hard to imagine" the current AI landscape without the contributions of Nvidia CEO Jensen Huang. Despite the logistical hurdles and rising costs, Altman observes a sharp divide in how the technology is being received: while investors remain increasingly bullish on AI's potential, he notes that journalists continue to express significant skepticism regarding the trajectory of the field.