Altman Called The Memory In Your PC “Quite Terrible.” Two Days Later He Met 50 Samsung Executives. Who Gets The World’s Memory Now?

🔍 Read the full analysis: Altman Called The Memory In Your PC “Quite Terrible.” Two Days Later He Met 50 Samsung Executives. Who Gets The World’s Memory Now? on ThorstenMeyerAI.com

Before you orderOffer from Amazon

Get the latest gadgets delivered free with Prime

• Fast, free delivery on millions of items
• Prime Video, Amazon Music and more included
• Member-only deals all year

Start your free Prime trial →Free trial for eligible customers · Cancel anytime

As an affiliate, we earn on qualifying purchases.Hide

OpenAI CEO Sam Altman met about 50 senior Samsung executives in Yongin, South Korea, on Oct. 31, two days after describing the allocation of memory in personal computers as “quite terrible” from a global supply perspective. Samsung said the gathering was part of an AI leadership program; no new memory-supply contract was announced, leaving the commercial purpose of the meeting unconfirmed.

Samsung said executives from its electronics, finance and biotech affiliates met Altman at the company’s Human Resources Development Center in Yongin. The session’s theme was “The Next 10 Years Shaped by AI: The Enterprise’s Choice.” Samsung described discussions about AI’s direction, its effect on business decisions and management, and markets it could open. The company did not publish an attendee list or disclose specific supply negotiations.

Bloomberg reported that the visit was intended to deepen cooperation on memory chips and AI infrastructure as shortages intensify. The Seoul Economic Daily reported that Richard Ho, OpenAI’s vice president overseeing its in-house AI chip work, accompanied Altman. That outlet also reported, citing industry sources, that an OpenAI accelerator could use Samsung’s next-generation HBM4 memory. Neither OpenAI nor Samsung confirmed that report.

Samsung’s account, as reported by Yonhap, said Altman told executives that companies gain an advantage not simply by adopting AI, but by using it to move faster, improve products and compete. The meeting followed an Oct. 2025 letter of intent between OpenAI and four Samsung companies. That earlier document said OpenAI’s memory demand was projected to reach up to 900,000 DRAM wafers per month; it was a projection, not a newly announced order at this meeting.

AI Demand Meets PC Memory

The visit puts a supply question at the center of the AI expansion: how much memory should go to data centers serving many users, and how much should remain in individual computers and other devices? Memory is a key input for AI servers, while Samsung is one of the world’s major memory producers. OpenAI’s reported interest in securing supply is significant because data-center growth can compete with consumer and business devices for manufacturing capacity.

Altman’s argument is that shared infrastructure could use memory more intensively than PCs that sit idle for much of the day. That is a policy and business argument, not a measured finding: the 5% figure he cited is his estimate, and the podcast discussion did not establish a universal usage rate for computers. His position also comes from the CEO of a company selling cloud AI services, which may benefit if organizations rent AI computing rather than run more workloads locally.

For buyers, the immediate concern described in the source material is not a confirmed shortage of retail products, but possible higher prices and changes to device specifications. TrendForce forecast conventional DRAM contract prices to rise 10% to 15% in the fourth quarter of 2026. It also warned that PC memory supply could fall in 2027 as more capacity shifts toward servers. Those are forecasts, not confirmed outcomes; if they materialize, manufacturers could face pressure to raise prices or offer less memory in some mainstream products.

Samsung’s results underline its exposure to the market. In the third quarter, the company’s chip division recorded 106.5 trillion won of Samsung’s 107.4 trillion won in operating profit, according to the source material. Samsung’s phone and network business posted an operating loss of 0.6 trillion won, partly amid cost pressures. The figures show the scale of the chip business in that quarter, but do not by themselves establish that AI demand caused the other division’s loss.

Products Worth Considering

Samsung 27 Odyssey OLED G5 (G50SF) QHD Gaming Monitor

AmazonView Latest Price

Samsung 27-inch Odyssey QD-OLED G8 4K 240Hz Gaming Monitor

AmazonView Latest Price

Samsung 32-inch Odyssey OLED G8 (G80SH) 4K QD-OLED Gaming Monitor

AmazonView Latest Price

As an affiliate, we earn on qualifying purchases.

From Podcast Remark to Yongin

Altman made the “quite terrible” remark in an episode of Cheeky Pint, the podcast hosted by Stripe co-founder John Collison, published on Oct. 29. Collison asked why high memory prices had not yet prompted more supply. Altman replied that “the invisible hand is failing right now” and argued that, from a global allocation perspective, a great deal of memory sits in personal computers while being used around 5% of the time. The source checked the passage against the recording; the 5% remains Altman’s estimate, not a measurement independently established in the episode.

The meeting took place on Oct. 31, about a year after Altman’s prior visit to Samsung, according to the Seoul Economic Daily. The relationship has also included the October 2025 letter of intent and a June 2026 OpenAI announcement that Samsung Electronics would make ChatGPT Enterprise and Codex available to employees in Korea and its global device division. The Korea Herald reported that Samsung executive chairman Lee Jae-yong visited OpenAI in San Francisco in July 2026.

Altman also met SK Group chairman Chey Tae-won in Seoul on Oct. 31, alongside executives from SK hynix, another major Korean memory maker, Yonhap reported, citing industry sources. The participants did not disclose what they discussed. These meetings come amid a broader effort by technology companies to strengthen supply relationships in South Korea; Bloomberg noted that Nvidia CEO Jensen Huang and AMD CEO Lisa Su had each visited the country twice since 2025.

“From a global allocation perspective, it is quite terrible that so much memory sits in people’s computers, being used 5% of the time.”

— Sam Altman, on the Cheeky Pint podcast, Oct. 29

Supply Talks Remain Undisclosed

No new supply deal was announced, and neither company disclosed volumes, prices or contract terms connected to the Oct. 31 meeting. Samsung described an AI leadership session, while Bloomberg characterized the visit as an effort to deepen memory and infrastructure cooperation. The public accounts do not establish whether the executives negotiated a specific purchase or reached any commercial understanding.

The reported presence of OpenAI chip chief Richard Ho and the possible use of Samsung HBM4 in an OpenAI accelerator also remain unconfirmed by both companies. The 900,000-wafer-per-month figure comes from a 2025 projection in a letter of intent; it should not be read as a current order or the outcome of this visit. It is also unclear how any future allocation decisions would affect memory available for PCs, phones or other devices.

The 2027 supply outlook is a forecast, not a settled account of future production. Actual availability and prices will depend on manufacturers’ capacity plans, demand and the pace of changes in the AI server market. The source material does not specify which PC models, product tiers or regions would face the greatest effects.

Watch for Capacity and Contracts

The next concrete indicators will be company disclosures about supply agreements, Samsung’s memory production plans and updated market forecasts. Any confirmed order, product commitment or HBM4 design announcement would clarify whether the discussions moved beyond the leadership session described by Samsung. For now, no public follow-up date or decision from the meeting has been announced.

Buyers and businesses can also watch quarterly memory-price updates and device specifications as manufacturers set plans for late 2026 and 2027. TrendForce’s projected 10% to 15% fourth-quarter increase in conventional DRAM contract prices and its warning about possible PC-memory supply declines are relevant signals, but they are forecasts rather than guarantees. The balance between data-center investment and memory for personal devices remains unresolved.

Key Questions

What happened at the Samsung meeting?

Samsung said about 50 senior executives from its electronics, finance and biotech affiliates met Sam Altman in Yongin on Oct. 31, 2026. Samsung described it as an AI leadership program covering technology, management and potential markets.

Did OpenAI and Samsung announce a memory deal?

No. No new supply contract, price or volume was announced. Bloomberg reported the visit aimed to deepen cooperation, but the companies did not publicly confirm a new agreement.

What did Altman mean by saying PC memory use was “quite terrible”?

He argued that memory installed in personal computers is often idle and could be used more intensively in shared data centers. His claim that it is used around 5% of the time was an estimate, not an independently verified measurement.

Could this make computers more expensive or reduce their memory?

That is a possible industry risk, not a confirmed result of the meeting. TrendForce forecast 10% to 15% growth in conventional DRAM contract prices for the fourth quarter of 2026 and warned PC memory supply could decline in 2027 as capacity shifts to servers. Those projections do not establish how prices or specifications for particular devices will change.

Source: ThorstenMeyerAI.com