The biggest players in AI are making a massive strategic pivot: moving directly into the physical infrastructure layer.
Why❓ Because production AI changed the economics.
The cost of an AI outcome is set across the entire path, and in a fragmented stack, every handoff between vendors adds cost, inefficiency, and complexity. 💸
👀Watch our sharing at @RaiseSummit to learn why controlling more of the stack matters for cost per outcome! 💡🌟
🎬 Catch the highlights on YouTube 👇
https://t.co/knDbo9jxu4
🧵 $GRRR: Why This AI Infrastructure Stock Could Rally 50-60% by September 1/ THE SETUP: Company: Gorilla Technology Group (NASDAQ: GRRR) Business: AI infrastructure & data center solutions Revenue FY2025: $101.4M (REAL, not promises) Growth: +35.7% YoY Operating Income: +$7.35M
5/ THE RISK: • Convertible notes dilution (2031) • Indonesia regulatory/political risk • Competition in AI data centers • Execution on NeutraDC Batam ramp-up But upside justifies risk.
🚨 @HIVEDigitalTech's @BUZZHPC Lands $350M AI Cloud Deal 🚨
⚡ $350M total contract value — five-year agreement with an investment-grade enterprise customer, adding ~$70M ARR
⚡ $180M total BUZZ ARR now — combining ~$35M live today with ~$145M in contracted ARR expected online through Q4 2026
⚡ 2,016 @nvidia Blackwell Ultra GPUs — deployed in GB300 NVL72 rack-scale systems at @Bell AI Fabric's Merritt, BC facility, running on 100% renewable hydroelectric power
⚡ ~10% upfront deposit — roughly $35M received against $185M in expected capex, with hive:native retaining full ownership of the infrastructure post-deployment
⚡ $500K daily HPC/AI revenue targeted — once the cluster deploys in Q4 2026, marking a major step toward hive:native's $200M year-end ARR target
WhiteFiber continues to amaze us all. Who knew they were approved by Jensen to deploy Veru Rubin VR200. I didnt. $WYFI is following CRWV & NBIS in every way as a powerful NeoCloud, but is also expanding its Colocation business (Flywheel to 100MW per year) and now “Redwood”. Get ready for $5B - $10B mcap (currently $1B mcap)
“We have entered into multi-year Cloud Services agreements representing more than $540M of contract value…expected to generate more than $200 million of annualized revenue when fully deployed.” @SamirTabar
$WYFI
BlackRock's 13F filing shows they increased their position in AmpliTech $AMPG to 522,988 as of June 30th, up significantly from 108,133 on March 31st.
Very bullish. Looks like institutional investors are slowly taking notice of the potential in $AMPG.
BREAKING: $NVDA AERIAL RAN GITHUB REPO SHOWS EVIDENCE THAT $AMPG 64T64R MIMO RADIO IS BEING INTEGRATED WITH AERIAL ON CODE-LEVEL!
Are you guys ready for the nerdiest AmpliTech alpha find so far? Credit for this goes to user @derducman!
A public pull request has been submitted to NVIDIA’s Aerial CUDA-Accelerated RAN repository titled:
“mMIMO support for Amplitech RU”
The submitted code is designed to make NVIDIA Aerial work correctly with AmpliTech’s 64T64R radio, including:
- 64-antenna mapping
- eCPRI communication fixes
- multi-MAC radio routing
- high-antenna-count channel estimation
- dynamic beamforming support
Why is this meaningful?
NVIDIA Aerial acts as the GPU-powered software brain of an AI-RAN base station, while AmpliTech provides the physical Massive MIMO radio hardware.
This suggests that AmpliTech’s radio is not merely being mentioned alongside NVIDIA, it is undergoing real, code-level integration with the Aerial platform.
The code was submitted by a developer from the Northeastern University/WINES Lab ecosystem. It has not yet been reviewed or merged by NVIDIA, but it is a legitimate technical development worth following.
What is the O-RAN CAT B 64T64R Massive MIMO RU?
This is the flagship product from Amplitech.
64T64R means it has 64 transmit and 64 receive paths, which makes it a Massive MIMO radio. It beamforms energy directly at users and multiplies network capacity.
In short, this is the most advanced and highest-value radio type in 5G. Almost no non-Chinese vendor can build a true CAT B Massive MIMO O-RU.
It remains the single biggest product gap in the O-RAN ecosystem, and that scarcity is the thesis. It is also exactly where AI-RAN lives, since AI beamforming needs 64T64R hardware.
..and now it is integrated to NVIDIA Aerial on code-level at the same moment when NVIDIA starts to post on their main channel about importance of 5G/6G.
Beyond bullish! 🔥🚀