"From the perspective of global resource allocation, having everyone hoard scarce DRAM in their own computers, idle most of the time, is a 'quite terrible' arrangement; memory should be concentrated in cloud data centers."
He really went there.
This Thursday, in an interview with the YouTube channel Cheeky Pint, OpenAI CEO Sam Altman drank beer throughout, chatting in a relaxed atmosphere about local inference of large AI models and the resulting problem of insufficient memory in home computers, and inadvertently dropped a bombshell.
As we know, the memory price surge triggered by the current global AI wave has already caused PC and smartphone prices to skyrocket, making them unaffordable for individual consumers. At the same time, capacity expansion by the major memory manufacturers will take a while. Just as the frontier AI companies are taking the lion's share of memory, they have already begun discussing whether memory in personal computers is being fully utilized.
Altman mentioned a hypothetical scenario in which people could run local AI on their own computers—computers they may already own, or ones they want to build because of the rise of large AI model technology. He thinks this is a bad idea.
"From the perspective of global resource allocation, having everyone leave the world's limited DRAM idle in their own computers with only 5% utilization is actually quite terrible," Altman said. "If you want to operate in an environment where this resource supply is extremely tight, then in the cloud you should maximize resource utilization. But for some reason, many people's intuition is exactly the opposite."
Today, OpenAI's main business is based on closed-source models running in large-scale data centers, and his defense of his company's business model has sparked a debate across the AI community.
This has also raised concerns among many: in the future, it may become even harder for us to own our own hardware or software, and we may only be able to consider renting cloud-based agents and computers.
Although the memory used by ordinary PCs, smartphones, and AI accelerator cards is not the same, their production competes for some upstream wafer capacity. Altman is discussing the allocation of global DRAM resources, and when it comes to who has the right to decide where memory capacity is used, most of the time it is naturally the customer placing the larger order.
Some of Altman's views are based on the assumption that data center construction will not slow down. He also assumes that memory used around the clock for AI data centers is more valuable than memory used for spreadsheets, web browsing, gaming, or even occasional local AI.
In fact, this interview was very long, and Altman also talked about the PHP vulnerability that sparked global panic over the past two weeks, sandboxing for training models, and AI research in mathematics.
Let's see what he mainly said:
AI Is Setting Off a Wave of Vulnerabilities
Regarding the enormous PHP vulnerability, Altman's judgment is: the ability of large models to discover vulnerabilities continues to improve rapidly, and the code surface on the internet is extremely vast. When people begin systematically using AI to scan all code, "we will find ourselves in a world where serious vulnerabilities far exceed expectations." He believes patching alone is not enough and that architectures must be rebuilt—routers, operating systems, and application layers all need to be redesigned for a world where "vulnerabilities will become increasingly intelligent." Two years ago, "rewriting the entire tech stack" still sounded crazy; now startups aiming to build entirely new stacks have already emerged.
He rarely offered industry self-criticism: "I don't think we have done a good job explaining to people why all the potential downside risks of this technology are worth it." He said the AI industry's narrative is bad sales pitch—promising to cure cancer and reduce costs on one hand, while depicting a future of unemployment, loss of control, and the need for strong regulation on the other. He reiterated that OpenAI insists on iterative deployment and confronting capability overhang, because "a few square miles of people in San Francisco" should not be making all the decisions.
When asked whether a "maximum-security prison" could be built for models, Altman gave a pessimistic answer: sandboxes are only effective to a certain extent, and monitoring (including chain-of-thought review and activation monitoring) can go further, but ultimately there must be a high degree of trust in alignment.He asked in return: when facing a model millions of times smarter than the smartest security researcher, one capable of discovering new science that humans do not yet understand, would you really dare to trust that sandbox?
He explicitly said he hates the term "agent swarms," because swarm evokes locusts, and insisted on calling them agent teams. He believes that the mutual checks and balances among teams correspond precisely to the mechanism of achieving robustness through pluralism, and that this is a safety direction worth pursuing.
The Intelligence Explosion and the Cost of Scientific Discovery
Altman sketched a curve: in the GPT-4 era, it was hard to effectively spend even $1 of compute on a single problem; chain of thought allowed models to "think hard," and now we can run an entire agent team in parallel tens of thousands of times. The usage gap between ordinary users and the most advanced users is constantly widening.
On how OpenAI allocates compute internally, he said employees enjoy great freedom, but high token consumption is never rewarded: "Being high on the token leaderboard is actually a bit embarrassing; it is certainly not a reason for promotion." He used the proof of the Navier-Stokes equations as an example: a single proof cost about $10 million, and during that period things "went out of control for a week," because at the time no one knew anything about the scaling laws of agent teams. But he expects that a year later, with smarter models and methods, it could be redone at 1/100 the cost.
He also rarely showed emotion: the situation at the time was truly incredible. On the weekend when a certain new model was launched internally, everyone threw all kinds of the toughest problems at the model, and it actually solved them one by one. People had not expected the model to be able to do these things at all. Together they witnessed many of the hardest scientific research problems in people's academic careers being solved one after another by this model, covering multiple different fields, and the scene was simply astonishing.
"For two nights I barely slept because I was so excited that I simply could not sleep. Many of the results we later published were actually those problems solved at that time."
At that time Altman felt extremely excited and happy, while also experiencing an unexpected sadness, melancholy, and a sense of "an era is coming to an end." It was not just nostalgia, but also empathy for human scientists' spirit of "hearing the Way in the morning, dying content in the evening." He insists that humanity deserves access to basic knowledge, but the industry owes practitioners in these fields a solution.
Although it is still too early to talk about superintelligence, Altman believes that a trillion slightly superhuman AIs are better than one with a trillion IQ:
“I would much rather intuitively have a trillion AIs of slightly superhuman intelligence join society than one AI of one trillion IQ join society.”
"Intuitively, I would rather have a trillion AIs with slightly superhuman intelligence join society than one AI with a trillion IQ join society."
His argument is: society's strong robustness comes from a large number of members checking and balancing and collaborating with each other; a "single-brain" world does not have this kind of resilience. But he emphasizes that this will not happen automatically and must be actively promoted.
Market adjustment no longer works in the AI era?
Regarding the global memory shortage, Altman introduced OpenAI's layout: self-developed chips (improving intelligence performance per watt), self-built data centers, robotics projects, and an even bolder concept, "self-replicating data centers," which would use AI to control robots to operate fabs, build turbines and solar panels, and construct data centers. He stated that "this is already within our grasp."
When asked why "the market regulates itself" has been so slow to materialize, he said current progress is quite bad, and gave the behavioral logic of memory manufacturers: this industry has long been a hard business of investing $20 billion in capex to earn $1.9 billion, and historically giants have repeatedly come close to bankruptcy due to overbuilding factories. Today each company can earn about $100 billion in free cash flow a year, but painful memories make them extremely conservative about expanding production. He thus asserted that "the invisible hand is failing now."
He distinguished two types of capabilities in the chip industry: for lithography, "it seems only one company on Earth knows how to do it," but chip design (the files ultimately sent to the fab) is already an excellent use case for AI.
Speaking about the real level of recursive self-improvement (RSI), Altman said models can now complete tasks such as "finding a 5% improvement in computational efficiency," but no one yet knows how to make AI discover a new architecture on the level of the transformer; what is missing is smarter models. He even predicted that society may need to agree for a period of time to limit the IQ of a single model, "and if future progress is insufficient, that will stem from this decision, not from AI's ability to do research."
He laid out his deflationary faith: the current five-year total cost of ownership for each GW of leading AI compute is about $70 billion, and it would be even better if it could drop to $20 billion, while models still need to improve performance by another 10x each year. The path to achieving this is to make energy cheaper and compress the stacked margins between "from sand to output tokens."
Regarding enterprises' anxiety about token budgets, he believes the more common problem today is not using AI too much, but not using it enough.
How should personal agents be priced?
Regarding the recently hot personal agent products (such as Muse), Altman's current judgment is that "the Macintosh moment has not yet arrived." People like these products, but they have not yet reached the point of "I can't imagine life without it." The winning form will resemble current products, but "it'll just work," reliable and able to handle more complex tasks. As for the cat-and-mouse game of computer use, he believes it will ultimately converge to the extent that "you have the right to let AI operate your own computer on your behalf."
Regarding the commercialization of agents, Altman believes enterprises cling to the AI pop-ups and HTML inside their own products, fearing the loss of customer relationships, but consumers want to have AI do everything on their behalf inside their main chat application. He believes the real moat comes from brand, product selection, and fulfillment networks, not HTML. He volunteered that he chooses Amazon Pay 10 times out of 10, even at a higher price, and praised Amazon's fulfillment network, while admitting that the partner platforms within ChatGPT are not good enough. He also called all web forms "fake work" and compared the present to the App Store era back then.
It is precisely for this reason that when designing pricing, OpenAI almost followed AWS in quoting by task and completely abandoning token-based billing, but encountered strong opposition from customers: "I need to know what I am paying for," even if they pay more, they want the understandable unit of tokens.
Reference content:
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
https://www.youtube.com/watch?v=iMsUcCYNKqY
This article comes from the WeChat public account "Machine Heart" (ID: almosthuman2014), author: Zenan
