Thread

Samsung is expected to more than double output of its HBM4 and HBM4E DRAM

562 points · 458 comments · giuliomagnifico

amelius · 1789928450 · focus · HN ↗
Will that be enough for AI's hunger?
1. GoToRO · 1789929801 · focus · HN ↗
  It will be just in time for when AI will run very well on consumer hardware and the need for data centers will collapse.
  1. EA-3167 · 1789930124 · focus · HN ↗
    Or even more amusing in time for the bubble to burst, I hope the big RAM makers end up holding the whole bag for that. Greedy bastards.
    bethekidyouwant · 1789930219 · focus · HN ↗
    Since you’re not greedy when there is a memory glut I’m sure you’ll be willing to pay extra to make it fair.
    EA-3167 · 1789930813 · focus · HN ↗
    Not at all, I’m going to relax with a cooling drink and watch the consequences of this unbelievably destructive and wasteful venture implode. I also adore the idea that it's on the customer to be "fair" to a company that dumped us a group in favor of chasing B2B money.
    Their choice, their consequences.
    pixl97 · 1789933296 · focus · HN ↗
    Internet is a fad, it will implode at any moment.
    EA-3167 · 1789933450 · focus · HN ↗
    In fact as a business opportunity it did just that because the massive investment in the mid-late 1990's was without anything like a connection to profitability or sustainable demand. It took years after that crash for the concept of e-commerce to begin both a recovery and evolution into what we see today.
    I expect something similar for AI. It's useful tech... just not very profitable tech, and certainly not to the tune of trillions of dollars worth of public demand. The promises of superintelligence, replacing everyone, and the rest of the hype will die with the companies who made the promises, but the tech will survive and thrive.
    pixl97 · 1789933916 · focus · HN ↗
    >The promises of superintelligence, replacing everyone, and the rest of the hype will die with the companies who made the promises
    The hype will die but there are many reasons why super intelligence and robots are a separate entity from said hype.
    Look, back a few decades and tell someone our gdp would be in the trillions and it's likely they'd have a hard time believing you. A huge portion of our products that we use day to day would be complete science fiction to them.
    All we're negotiating at this point is the timescale it will take.
    card_zero · 1789937577 · focus · HN ↗
    Unpredictable things happen routinely, therefore these specific unpredictable things will happen?
    pixl97 · 1789939653 · focus · HN ↗
    These things aren't really unpredictable. Work long enough on a robot with more dexterity and you will get just that. The timetable on when it happens is a bit more up in the air, economic downturns and wars can have huge impacts on it.
    But in my mind robots and superintelligence are inevitable unless there is a massive setback in humanity before then. Nature already did it once. We're not inventing something totally new. Add in every new invention and bit of intelligence we build on and actualize pushes us that much closer to the goal. Information technology allows this to speed up even further with easy sharing of information and testing.
    Superintelligence is not like faster than light travel. We have many frameworks that show us FTL is impossible. I don't believe there is a single widely accepted framework that shows there is some limit to intelligence and we're near it. Intelligence isn't even a singular thing. My calculator is a super intelligent adder compared to me. It would seem highly improbable that somehow nature random walked into making brains the most efficient general intelligence device in all dimensions and scales.
    usef- · 1789944559 · focus · HN ↗
    I think they're talking about the past: memory has often been boom/bust. Memory manufacturers lose money during gluts, and did so long before AI. People benefited from cheap prices at that time.
    glub · 1789940484 · focus · HN ↗
    Nearly the entire reason we're in this mess is because RAM/SSD/HDD makers are terrified of AI bubble bursting.
    Long-term contracts, not building new fabs - they're in full on hedging mode right now.
    leoc · 1789948233 · focus · HN ↗
    Seemingly the AI bulls are either not sufficiently optimistic, or not sufficiently wealthy(?!), to build new fabs as joint ventures with the manufacturers in which they agree to assume most of the downside risk?
    no-name-here · 1789954716 · focus · HN ↗
    They signed long term contracts for the hardware delivery over years. AI investment is approaching 1 trillion per year and expected to grow to well over 1 trillion per year. [1]
    Even over their many years, projects like the Manhattan Project, the Apollo Program, or the U.S. Interstate Highway System never added up to that. [2]
    Is your argument that there is insufficient AI spending at present as they aren’t also taking on building their own fabs?
    [1] <a href="https://www.pwc.com/gx/en/news-room/press-releases/2026/global-investment-in-ai-infrastructure.html" rel="nofollow">https://www.pwc.com/gx/en/news-room/press-releases/2026/glob...
    [2] <a href="https://www.aljazeera.com/news/2026/2/19/visualising-ai-spending-how-does-it-compare-with-historys-mega-projects" rel="nofollow">https://www.aljazeera.com/news/2026/2/19/visualising-ai-spen...
    m4rtink · 1789949449 · focus · HN ↗
    If the long term contact are with any AI companies, then good luck getting anything back in bankrupcy proceedings!
    m4rtink · 1789949359 · focus · HN ↗
    Looking forward to see them burn.
    ThrowawayR2 · 1789960550 · focus · HN ↗
    If the AI bubble bursts, the DRAM manufacturers switch back to making conventional DRAM and pent-up business and consumer demand results in a surge of purchases, cushioning the fall considerably.
    EA-3167 · 1790010131 · focus · HN ↗
    Look into the time frames a bit more, this isn’t a switch they flip. If the bubble bursts today it will be years before those downstream effects add up.

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.