OpenAI Moves to Acquire Neptune as It Expands AI Training Capabilities 
Microchip Technology Boosts Q3 Outlook on Strong Bookings Momentum 
Visa to Move European Headquarters to London’s Canary Wharf 
Sam Altman Reportedly Explored Funding for Rocket Venture in Potential Challenge to SpaceX 
Wikipedia Pushes for AI Licensing Deals as Jimmy Wales Calls for Fair Compensation 
AI-Guided Drones Transform Ukraine’s Battlefield Strategy 
USPS Expands Electric Vehicle Fleet as Nationwide Transition Accelerates 
Hikvision Challenges FCC Rule Tightening Restrictions on Chinese Telecom Equipment 
Trump Administration to Secure Equity Stake in Pat Gelsinger’s XLight Startup 
UPS MD-11 Crash Prompts Families to Prepare Wrongful Death Lawsuit 
Intel Boosts Malaysia Operations with Additional RM860 Million Investment 
Anthropic Reportedly Taps Wilson Sonsini as It Prepares for a Potential 2026 IPO 
ExxonMobil to Shut Older Singapore Steam Cracker Amid Global Petrochemical Downturn 
Michael Dell Pledges $6.25 Billion to Boost Children’s Investment Accounts Under Trump Initiative 
Firelight Launches as First XRP Staking Platform on Flare, Introduces DeFi Cover Feature 
Norway’s Wealth Fund Backs Shareholder Push for Microsoft Human-Rights Risk Report 
Tech moguls Elon Musk and Larry Ellison begged Nvidia CEO Jensen Huang over dinner for more AI GPUs, as Oracle plans a massive supercluster powered by Nvidia technology, signaling the escalating race in artificial intelligence.
Elon Musk and Larry Ellison Begged Nvidia’s Jensen Huang for GPUs
Oracle co-founder Larry Ellison, speaking during the most recent earnings call, acknowledged (via Barron's) that he had to grovel to Nvidia CEO Jensen Huang to get the latest graphics processing units (GPUs) for his company.
“In Nobu Palo Alto, I went to dinner with Elon Musk, Jensen Huang, and I would describe the dinner as me and Elon begging Jensen for GPUs. Please take our money; no, take more of it. You’re not taking enough of it; we need you to take more of our money, please,” Ellison pleaded with the other party throughout the call. “It went okay; it worked.”
Looking back, it appears that Ellison and Oracle got a lot out of this dinner.
More powerful than Musk and xAI's Memphis Supercluster, which presently contains 100,000 Nvidia H100 AI GPUs, the company's upcoming Zettascale AI supercluster, made of 131,072 Nvidia GB200 NVL72 Blackwell GPUs, would give 2.4 ZettaFLOPS of AI performance.
Mobile Generators to Support Oracle’s AI Projects Until Reactors Are Online
Oracle has already obtained permission to construct three modular nuclear reactors to supply its buildings with electricity, as its AI plans require a tremendous amount of power. But since it will probably take a while for data centers to have nuclear reactors installed, the business may, for the time being, follow Musk's example and use huge mobile generators to increase the local power supply if necessary.
Although it is smaller than other large IT businesses that provide data center services, Oracle Cloud Infrastructure (OCI) has a clear edge over Amazon Web Services (AWS), Microsoft Azure (Azure), and Google Cloud (Google Cloud).
Offline Servers Cater to Oracle’s Security-Conscious Clients
When it comes to meeting the unique needs of certain clients, Tom's Hardware claims that OCI offers superior flexibility. For the most security-conscious customers, it goes so far as to offer offline servers that operate on its own networking infrastructure.
Oracle may be little, but it is making a big splash with its AI initiatives.
Ellison, echoing Anthropic CEO Dario Amodei, predicts that the training of frontier AI models due out in the next three years will cost $100 billion. Plus, it appears that OCI aspires to be a frontrunner in the artificial intelligence processing industry.
"Someone’s going to be better at this than anybody else, and multiple people are trying, and there is a race," Ellison stated during the call. Finally, he chimed in, saying, "Getting there first is a big deal."
