- Generative AI Art
- Posts
- xAI and Microsoft's massive AI Datacenters
xAI and Microsoft's massive AI Datacenters
PLUS: OpenAI and Jony Ive's new hardware and NVIDIA's $900M talent grab
A race to build the world's most powerful AI datacenters is underway, led by Microsoft and xAI. The massive investment wave signals a new era focused on creating the immense computational power required to train the next generation of AI models.
This new infrastructure is for models like xAI's upcoming Grok 5, which Elon Musk claims could "reach AGI." With companies betting billions on compute, is access to this level of processing power the final ingredient needed to unlock more advanced AI?
Today in AI:
xAI and Microsoft’s gigawatt-scale data centers
OpenAI and Jony Ive’s new hardware venture
NVIDIA’s $900M networking talent grab
PRESENTED BY DESELECT
Enable non-technical marketers to run sophisticated segmentation in Salesforce Marketing Cloud.
Empower your marketing team with DESelect Segment, requiring no SQL skills. Easily create sophisticated audience segments directly within Salesforce Marketing Cloud using intuitive drag-and-drop tools.
Boost your campaign precision and save time with automated, error-free segmentation. Discover the future of marketing efficiency today.
What’s new? Microsoft and xAI are building two of the world's most powerful AI datacenters, signaling a massive investment wave to power next-generation models.
What matters?
Microsoft's new Fairwater campus in Wisconsin is being called the world's most advanced AI datacenter, set to house hundreds of thousands of GPUs.
Not to be outdone, xAI is building Colossus 2, the world’s first “gigawatt scale” data center, equivalent to the power output of a nuclear plant.
The immense compute is for training new models, with Elon Musk claiming the upcoming Grok 5 could “reach AGI” once it begins training in a few weeks.
I now think @xai has a chance of reaching AGI with @grok 5. Never thought that before.
— Elon Musk (@elonmusk)
6:36 AM • Sep 17, 2025
Why it matters?
These projects show that access to massive computational power is now seen as the key ingredient for creating more capable AI. The companies that build the most effective infrastructure today are positioning themselves to lead the entire AI industry tomorrow.
AI TOOLS
What’s new? OpenAI is making a major move beyond software, teaming with Jony Ive to build a new line of AI-native hardware. This venture aims to create the next generation of devices designed from the ground up for artificial intelligence.
What matters?
The new venture is aggressively acquiring talent, having already poached over 24 engineers from Apple with stock grants reportedly exceeding $1M.
Early reports suggest the team is exploring a range of devices, including smart glasses, voice recorders, and other wearable pins.
This is not just an experiment; the project has a target release window for its first products, aiming for late 2026.
Why it matters?
This partnership signals a critical shift for AI, moving it from our screens into dedicated physical devices. Bringing Jony Ive's design expertise to the table could produce the first truly intuitive AI hardware, setting the standard for the next era of consumer technology.
PRESENTED BY ROKU
It’s go-time for holiday campaigns
Roku Ads Manager makes it easy to extend your Q4 campaign to performance CTV.
You can:
Easily launch self-serve CTV ads
Repurpose your social content for TV
Drive purchases directly on-screen with shoppable ads
A/B test to discover your most effective offers
The holidays only come once a year. Get started now with a $500 ad credit when you spend your first $500 today with code: ROKUADS500. Terms apply.
What’s new? NVIDIA reverse-acquihired the top management from networking startup Enfabrica for over $900 million. This unconventional move bypasses a traditional acquisition to directly secure critical talent.
What matters?
This move directly addresses a major bottleneck in building massive AI clusters: GPU connectivity. As systems grow, the "plumbing" that allows thousands of GPUs to communicate efficiently is as important as the chips themselves.
The talent grab is a clear play to support NVIDIA's biggest customers, who are building the next generation of gigawatt-scale data centers and need highly specialized networking solutions.
By bringing Enfabrica’s leadership in-house, NVIDIA is solidifying its ecosystem and defending its market dominance against competitors working to solve the high-speed networking problem.
Why it matters?
Spending nearly a billion dollars on a leadership team signals that high-speed interconnects are now a top priority for advancing AI. This shows the next battleground for AI infrastructure is about designing the entire data center, not just making faster chips.
Everything else in AI
Huawei unveiled DeepSeek-R1-Safe, a model trained specifically to block politically sensitive outputs, reportedly achieving an 83% “security defense” score to align with China’s regulatory standards.
Khazna expands its global AI infrastructure footprint with three new AI-optimized facilities in the UAE and plans for a massive 1GW expansion across Europe and the Middle East.
Researchers revealed the “Shadow Leak” exploit, a vulnerability in ChatGPT’s Deep Research feature that could be hijacked to steal sensitive information from users’ connected Gmail accounts.
Stallone pitched the idea of using AI to de-age himself to star as a teenage Green Beret in a potential Rambo origin story prequel.
Essential AI Guides - Reading List:
Let us know!
What did you think of today's email?Before you go, please give your feedback to help us improve the content for you! |
Work with us
Reach 100k+ engaged Tech Professionals, Engineers, Managers and decision makers. Join brands like MorningBrew, HubSpot, Prezi, Nike, Ahref, Roku, 1440, Superhuman, and others in showcasing your product to our audience. Get in touch now →