LightNews LightNews
chiper007.bsky.social
@chiper007.bsky.social
10 followers 31 following 0 posts
Posts Replies Media Videos
Reposted
tahjnaar.bsky.social
I disagree. I want to be able to vote on collapsing the United States instead.
January 14, 2026 at 6:21 AM
Reposted
edzitron.com
The real OpenAI scandal is what's *not* happening - Broadcom said it expects "no revenue" in 2026 from its supposed 10GW chips deal, and AMD, Samsung and SK Hynix, despite signing massive deals, haven't remotely changed earnings. These announcements were a sham.
www.wheresyoured.at/premium-how-...
Samsung and SK Hynix's OpenAI's DRAM Deals Are Bullshit Back in October, Reuters reported that Samsung and SK Hynix had "signed letters of intent to supply memory chips for OpenAI's data centers," with South Korea's presidential office saying that said chip demand was expected to reach "900,000 wafers a month," with "much of that from Samsung and SK Hynix," which was quickly extrapolated to mean around 40% of global DRAM output. That's a lot of fucking RAM! One would think that both SK Hynix and Samsung would have it all over their respective earnings calls. And you'd be dead wrong. On SK Hynix's October 29 2025 earnings call, weeks after the announcement, SK Hynix was asked a question about High Bandwidth Memory growth by SK Kim from Daiwa Securities: Kim: Thank you very much for taking my question. It is on demand. Now, there have been a series of announcements of GPU and ASIC supply cooperation between Big Techs and AI companies, fueling expectations of further AI market growth. Then, against this backdrop, what is the company's outlook on HBM demand growth, as well as a broadening of the customer base? SK Hynix: Thank you for the question. Now, with upward adjustment in Big Tech's CapEx and increased investment by AI companies, the HBM market, even by a conservative estimate, will keep growing at an average of over 30% for the next five years. I will point to our recent LOI with OpenAI for large-scale DRAM supply as an example of the very strong market demand for AI, as well as the need to secure AI memory based on HBM more than anything else when developing AI technology. This is the only mention of OpenAI. Otherwise, SK Hynix has not added any guidance that would suggest that its DRAM sales will spike beyond overall growth, other than mentioning it had "completed year 2026 supply discussions with key customers." There is no mention of OpenAI in any earnings presentation. On Samsung's October 30 2025 earnings call, Samsung mentioned the term "DRAM" 18 times, and neither mentioned OpenAI nor any letters of intent. In its Q3 2025 earnings presentation, Samsung mentions it will "prioritize the expansion of the HBM4 business with differentiated performance to address increasing AI demand." Not a single fucking mention of OpenAI. This is a deal that allegedly will divert 40% of the global supply of RAM to one company, yet neither of the companies set to sell all that RAM appear to be talking about the deal or giving Wall Street any guidance about how much money it'll make. Furthermore, what would OpenAI do with all the RAM exactly? Where is it going? Is it going to get GPUs without RAM on them and then put the RAM on? How will it do that? Where will the RAM be kept? These are unfinished wafers of RAM, right? Making RAM is complex (here's a guide by GamersNexus) business, and even if the wafers arrive in the latter stages — trimmed, formed, with burn-in testing complete — is OpenAI going to manufacture the finished product? How? Where? On-site at the data centers? Or is the RAM going to be fully complete by the time it gets to the data centers? It's almost as if this deal is complete nonsense used to boost OpenAI, Samsung and SK Hynix's profile without anybody actually having to do anything! Why does nothing happen every single time that OpenAI would theoretically have to do something or pay somebody a dollar? I think it's because OpenAI doesn't actually have any cash it can afford to hand to anybody other than its compute providers.
December 22, 2025 at 5:55 PM
Reposted
edzitron.com
Based on everything I've read, every single OpenAI deal is bullshit - AMD, NVIDIA, Broadcom, SK Hynix and Samsung are acting as if they will never get paid by Sam Altman. And no, OpenAI isn't causing the global RAM shortage.
edzitronswheresyouredatghostio.outpost.pub/public/promo...
Are Any of OpenAI's Big Deals Actually Real? The Financial Times reported back in September that OpenAI was "set to produce its first artificial intelligence chip [in 2026] with US semiconductor giant Broadcom," and reports about the chip go back to as early as October 2024. Per Reuters: OpenAI has been working for months with Broadcom to build its first AI chip focusing on inference, according to sources. Demand right now is greater for training chips, but analysts have predicted the need for inference chips could surpass them as more AI applications are deployed. So this supposed OpenAI chip has existed for nearly two years, yet nobody seems to be able to explain what it is, how it will be paid for, or when it will be delivered. Per OpenAI's October announcement: OpenAI and Broadcom have long-standing agreements on the co-development and supply of the AI accelerators. The two companies have signed a term sheet to deploy racks incorporating the AI accelerators and Broadcom networking solutions. Okay, a term sheet. Cool. Great… does that mean anything? Because I cannot find any evidence that OpenAI and Broadcom are actually building anything or that OpenAI has paid it a dime to do so. The Wall Street Journal says that the deal "will be worth multiple billions of dollars" according to people familiar with the matter, yet these multiple billions of dollars do not appear to be part of Broadcom's alleged (and "dynamic") $73 billion in backlog which will allegedly be paid over the next 18 months. In simpler terms, I can find no tangible evidence that OpenAI and Broadcom are taking any serious steps to manufacturing a chip, nor can I find any evidence that OpenAI has sent Broadcom money. Broadcom is not counting its OpenAI revenue in its 2026 guidance, nor is it discussing any kind of installation of these chips, nor has OpenAI shown much evidence that it will even build a data center to put them in. Similarly, I cannot find any evidence of anything new happening with Ope… Yet when you dig into AMD's 10-Q... As of September 27, 2025, the aggregate transaction price allocated to remaining performance obligations under contracts with an original expected duration of more than one year was $279 million, of which $139 million is expected to be recognized in the next 12 months. The revenue allocated to remaining performance obligations does not include amounts which have an original expected duration of one year or less. Huh. Where's all that OpenAI revenue? Its guidance didn't dramatically change, nor did there appear to be any spikes in operating expenses, investments, or...anything. Much like Broadcom, AMD is acting like somebody that doesn't think OpenAI will drive any revenue of any kind. AMD raised guidance by 35% over the next five years. How does that fit in with a deal that was, to quote Jean Hu, EVP, CFO and Treasurer at AMD, worth "tens of billions of dollars in revenue"? AMD's trailing 12-month revenue is $32 billion. "Tens of billions of dollars" would surely lead to more than a 35% boost (an increase of $11.2 billion or so) in the next five years? Sounds like it doesn't exist to me! Oh, that $100 billion deal with NVIDIA? The one that allegedly would see NVIDIA wire OpenAI $10 billion in October, if you listened to CNBC? It's not real! It's a fucking Letter of Intent! Per NVIDIA's latest 10-Q: In the third quarter of fiscal year 2026, we entered into a letter of intent with an opportunity to invest in OpenAI. No investment amount, no further details. NVIDIA isn't investing a dime.
December 19, 2025 at 8:33 PM
Reposted
gedsperber.bsky.social
I've written a short comment regarding this new MIT study that is being passed around and why it is horseshit. "11.7% of US workforce is being replaced by AI" my ass. Thought that might be of interest. Also - thanks for all your work, you are one of the best resources out there.
AI isn't coming for your job: Melting the Iceberg
Why the new MIT Study speaking of 11.7% automation is horseshit and why the way it's being reported is even worse
open.substack.com
November 30, 2025 at 5:56 PM