> ## Content Index > Fetch the complete content index at: https://www.implicator.ai/llms.txt > Use this file to discover other available public pages before exploring further. # Idle Memory: Altman Eyes the RAM Sitting in Your PC - URL: https://www.implicator.ai/when-talking-about-local-inference-sam-altman-says-supply/ - Published: 2026-10-31T12:09:43.000Z - Updated: 2026-10-31T12:53:04.000Z - Description: OpenAI's chief says the memory in home computers would do more good in the cloud. The idle-time figure is his own, and data centers running closed models are his company's main business. - Author: Marcus Schuler - Tags: AI News, #auto, #home-companion OpenAI CEO Sam Altman [said this week](https://www.theverge.com/ai-artificial-intelligence/1023820/sam-altman-ram-shortage-cloud-servers?ref=implicator.ai) that memory is better used in data centers than in people's home computers. On Stripe cofounder John Collison's [Cheeky Pint podcast](https://gizmodo.com/sam-altman-all-the-ram-should-belong-to-me-so-i-can-charge-people-to-use-it-2000833412?ref=implicator.ai), he put the share of time that DRAM in a personal computer gets used at 5%. The remark landed while DRAM is already expensive because of heavy demand from AI data centers, OpenAI's among them. What Changed - OpenAI CEO Sam Altman said on the Cheeky Pint podcast that DRAM in personal computers gets used 5% of the time. - He said maximum utilization in the cloud is what a supply-constrained market needs. - Samsung and SK Hynix have long-term deals with cloud operators and projects like OpenAI's Stargate that reserve much of the world's RAM output. - A Tom's Hardware editor wrote that Altman is defending OpenAI's business model of closed models running in data centers. AI-generated summary, reviewed by an editor. [More on our AI guidelines](https://www.implicator.ai/about/). ## What Altman said The comment came about [29 minutes into the interview](https://www.tomshardware.com/tech-industry/data-centers/openai-ceo-says-its-quite-terrible-for-your-pc-to-have-ram-sam-altman-says-data-centers-offer-better-utilization?ref=implicator.ai), in a discussion of local inference, meaning AI models run on people's own computers. "A lot of people want local inference," Altman said. People assume, he said, that open-source models running on their own machines mean "we're not going to need all the stuff in the data center." "But 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," he said. "You want maximum utilization in the cloud if you're trying to deal with this very supply-constrained environment. And somehow many people's intuition is the opposite of that." He framed the chip limits as a common problem. "Given that the world is going to be limited in its silicon capacity, we all have a shared interest," he said. In the same conversation, Altman said OpenAI is [designing its own custom AI processor](https://frvr.com/blog/news/openai-ceo-sam-altman-says-its-terrible-to-have-everybody-have-a-bunch-of-the-worlds-limited-supply-of-dram-instead-of-shoving-it-all-in-datacentres/?ref=implicator.ai), called Jalapeno, and that AI is designing the chip. FREE AI BRIEFING · WEEKDAYS Track where the world's memory supply goes. Get the AI stories shaping the day, with concise context from San Francisco. The briefing takes about five minutes and arrives at 4:45 a.m. Pacific, 7:45 a.m. Eastern. Email address Keep me informed Check your inbox for the confirmation link. Free. No hype. Unsubscribe anytime. ## Where the memory goes Samsung and SK Hynix have long-term deals with cloud service operators and with projects like OpenAI's Stargate. Those deals have reserved much of the world's RAM output over the next few years. The billions those buyers are willing to spend on chips are why laptops and phones are getting more expensive right now. Micron has announced its departure from consumer memory production. The company called the needs of AI data centers essential compared to consumer use. Intel shares [plunged almost 30 percent on Friday](https://futurism.com/artificial-intelligence/sam-altman-hates-local-ai-models-computers?ref=implicator.ai), the worst stock collapse in 52 years. Analysts said a massive slowdown in the PC market could be the cause. ## OpenAI's stake in the argument OpenAI does have some open-source models, but its primary business is closed-source models that run in data centers. Andrew E. Freedman, a senior editor at Tom's Hardware, wrote that Altman is "really defending the company's business model." Local AI could one day mean people pay less for subscriptions to OpenAI's models, Freedman added, if they have the power to run them at home. Gizmodo put OpenAI's infrastructure rollout at $750 billion through 2030 and said the cost of that rollout is driving up DRAM prices. "Altman isn't describing the optimal distribution of resources so much as the distribution of resources optimal to him," Gizmodo wrote. Know someone who'd find this useful? [✉️ Email it to a friend in one click](mailto:?subject=A%20newsletter%20I%20think%20you%27d%20like&body=This%20is%20one%20of%20maybe%20three%20newsletters%20I%20actually%20read.%20The%20rest%20just%20pile%20up%2C%20unread%2C%20judging%20me.%0A%0AAnd%20yes%2C%20this%20email%20mostly%20wrote%20itself%2C%20which%20is%20a%20little%20on%20the%20nose%20for%20an%20AI%20newsletter.%20Doesn%27t%20make%20it%20wrong.%20implicator.ai%20is%20good.%0A%0ASubscribe%20free%3A%20https%3A%2F%2Fwww.implicator.ai%2Fsubscribe%2F%3Futm%5Fsource%3Dnewsletter%26utm%5Fmedium%3Dforward%26utm%5Fcampaign%3Demail%5Fforward), or they can [subscribe free here](https://www.implicator.ai/subscribe/?utm%5Fsource=newsletter&utm%5Fmedium=forward&utm%5Fcampaign=forward%5Fto%5Fcolleague). ## What the 5% leaves out The 5% figure is Altman's own, offered in an interview over beers. Freedman pointed to two assumptions under it. One is that data center buildouts won't slow down. The other is that RAM used around the clock in AI data centers is more worthwhile than RAM used for spreadsheets, web browsing, gaming and the occasional local AI job. Running models at home has grown with the emergence of open-weight models from China. Altman's comments drew the ire of the PC gaming community. "He seems to not be able to comprehend the idea of people wanting to own their own computers," one Reddit user wrote. "Like it's an alien concept to him." The memory makers do not expect relief soon. Samsung expects the RAM shortage to last until at least 2028\. SK Hynix says memory shortages will last through 2030. Frequently Asked Questions What did Sam Altman say about RAM in home computers? On the Cheeky Pint podcast, Altman said that from a global allocation perspective it is quite terrible for everybody to have the world's limited supply of DRAM sitting in their own computer and being used 5% of the time. He said maximum utilization in the cloud is what a supply-constrained environment needs. Where does the 5% figure come from? The figure is Altman's own, offered in an interview over beers on a podcast. Tom's Hardware editor Andrew E. Freedman noted the argument assumes data center buildouts won't slow down and that RAM used around the clock in AI data centers is more worthwhile than RAM used for everyday tasks. Why are memory prices high? DRAM is expensive because of heavy demand from AI data centers. Samsung and SK Hynix have long-term deals with cloud service operators and projects like OpenAI's Stargate that have reserved much of the world's RAM output over the next few years. Does OpenAI benefit from the argument? OpenAI's primary business is closed-source models that run in data centers. Freedman wrote that Altman is really defending the company's business model, and that local AI could one day mean people pay less for subscriptions to OpenAI's models. How long will the memory shortage last? Samsung expects the RAM shortage to last until at least 2028\. SK Hynix says memory shortages will last through 2030. AI-generated summary, reviewed by an editor. [More on our AI guidelines](https://www.implicator.ai/about/). [Alphabet's AI bill stings shareholders; Microsoft leans on OpenAI revenueIMPLICATOR .ai Morning Briefing · From San Francisco Thursday, October 29, 2026 9 stops From San Francisco 1 The Editorial Marcus here. Today's technology bills haThe Implicator![](https://www.implicator.ai/content/images/2026/10/20261028-191315-nl_cloud_capex_hall.webp)](https://www.implicator.ai/alphabet-cloud-cash-microsoft-openai-revenue/) [PHP Fixes a Memory Bug Flagged in 2024 After AI Turned It Into a Working ExploitOn Aug. 11, 2026, Thai Duong, chief executive of the security firm Calif, called Volker Dusch, the PHP Foundation’s AI Security Engineer in Residence. A researcher had found a way to take over serversThe Implicator![](https://www.implicator.ai/content/images/2026/10/20261021-213137-php_patch_offline.webp)](https://www.implicator.ai/php-patches-ai-found-memory-flaws/) [Which AI memory tool is best for your coding agent?The Implicator![](https://www.implicator.ai/content/images/2026/10/20261006-ai-memory-tool-coding-agent.webp)](https://www.implicator.ai/claude-mem-leads-ai-coding-memory-plugin-ranking/)