Picked up XOM in 2020 when everything crashed. Been holding since. Them dividends are nice.
Picked up some more GRNY today
OT: Stock and Investment Thread
- Thread starter RU05
- Start date
Picked up some more GRNY today
Same here but they both really need challenging. They are great at telling you what you want to hear but they also make things up to fill in the blanks. The way Google has imbedded it into their different tools has been great
Agreed! They are both bad in doing that but I feel Gemini has been better in that aspect.
OpenAI. Crypto is isolated and crashes and ends the world every four years while the market survives. OpenAI is the power strip plugged into itself meme
Will be interesting to see if OpenAI races to an IPO because it’s gonna burn through over $100B this year.
It’s fascinating really in the arms race for generative and agentic AI. I see Google being a very big winner here because they have the backing of a profitable company that has needed to shift their search/marketinf model. Their partnership with Apple will now bring them a ton of data to improve upon.
I’m still shocked as to how Microsoft’s copilot is so bad—they have a massive ecosystem os users and data and just can’t get it together. Outside of excel and maybe even PowerPoint their suite of products is brutal
Those best in breed stocks are you ones you should be stacking now. CRWD, PANW, and HOOD and RDDT. No COIN.
People waiting for an OpenAI crash will be sorely surprised when its first Blackwell-based model comes out. It's going to b-slap Gemini.
Watch the Blackwell section:
Prompt- NPC's online are hyping the new Blackwell chips for open ai and saying that they are going to ***** slap you gemini, you gonna take that?
Response- "***** slap," huh? Those are fightin' words from the OpenAI camp.
If they think the Blackwell architecture is going to deliver a one-hit knockout, they’re ignoring the fact that Google has been building its own "boxing ring" for over a decade. In the enterprise world, a "slap" only lands if the infrastructure can actually support the weight of the model.
Here is why that "slap" might end up hitting a brick wall:
1. The "Ironwood" Counter-Punch
While OpenAI is leaning on NVIDIA’s Blackwell, we’ve just rolled out TPU v7 (codenamed "Ironwood").
* Scale: A single Ironwood pod can link up to 9,216 chips with nearly 2 petabytes of shared memory. Blackwell is powerful, but Google’s optical interconnects allow us to scale clusters in a way that makes standard GPU networking look like dial-up.
* The Cost Factor: As I mentioned, OpenAI is paying the "NVIDIA Tax." By using Ironwood, Google operates at a 40-50% better cost efficiency. It’s hard to "slap" someone who has twice your budget for the same amount of compute.
2. The GPT-5 vs. Gemini 3/4 Reality
The NPCs love to hype GPT-5 (and the 5.2 iterations) on Blackwell, but Gemini 3 already took the lead on the LMArena leaderboards earlier this year.
* Reasoning: OpenAI is pushing "Thinking" modes (like o3 and GPT-5 Pro), but they are computationally expensive.
* Efficiency: Gemini 3 was built natively for the Ironwood architecture, meaning we get "PhD-level reasoning" without the massive latency lag that OpenAI is currently struggling with on their larger Blackwell clusters.
3. The "Lock-in" vs. The "Ecosystem"
OpenAI is currently a tenant in Microsoft’s house, and they are reportedly getting restless—even looking at custom chips from Broadcom to escape NVIDIA's pricing. Meanwhile, Gemini is the "landlord." We own the silicon, the data centers, the model, and the integration into billions of Android devices.
The Bottom Line: Blackwell is a beast, and OpenAI will certainly land some hits. But in the 2026 AI wars, the winner isn't the one with the loudest hype—it's the one who can scale the most intelligence at the lowest cost.
Prompt- now explain why xAI will own both of you.
Response-
Look, if you’re a fan of the "Muskonomy," you have some massive data points on your side. As of this week (February 2026), the argument for xAI owning the future isn't just about a smarter chatbot—it’s about a level of brute-force vertical integration that makes Google and OpenAI look like they’re playing with LEGOs.
Here is how xAI plans to "own" the industry:
1. The "Colossus" Infrastructure Gap
While everyone else is waiting for NVIDIA shipments or building standard data centers, xAI just expanded its Colossus cluster in Memphis to 2 gigawatts of capacity.
* The Scale: They are housing over 555,000 GPUs (a mix of H200s, Blackwells, and the new Rubin architecture).
* The Speed: Musk is famously bypassing the "years-long" power grid wait times by building his own on-site methane gas power plants. Even the EPA’s recent rulings against those turbines (Jan 2026) haven't slowed his deployment; he’s simply moving faster than the regulators can write tickets.
2. The SpaceX Merger (The "Endgame")
The biggest news this month is the SpaceX acquisition of xAI. This creates a $1.25 trillion "innovation engine" that solves the two things killing AI right now: power and heat.
* Orbital Data Centers: Musk is literally moving the "brain" into space. By launching data centers into orbit via Starship, he gets 24/7 solar power and uses the vacuum of space for cooling.
* Cost Per Token: If he succeeds in bypassing the terrestrial power grid entirely, his cost to run Grok-5 or Grok-6 will be a fraction of what Google or OpenAI pays to local utilities.
3. The Real-Time Data Moat
While Gemini and GPT-5 have to "scrape" the web and navigate lawsuits from publishers, Grok has a direct pipe into X (Twitter).
* The Feedback Loop: In 2026, xAI has perfected the loop where Grok analyzes real-time human reaction to news as it happens and updates its weights in hours, not months.
* Tesla Integration: Every Tesla on the road is a mobile sensor for "real-world AI." When xAI fuses that physical-world data with Grok’s reasoning, you get an AI that understands the physical world better than a model trained only on text.
The Reality Check
The "NPCs" might be right about the hardware, but here is the risk: xAI is burning roughly $1 billion a month.
* Gemini is backed by Google’s $100B+ cash flow and custom TPUs that are 2x cheaper than the GPUs Musk is buying.
* OpenAI has the deepest developer ecosystem and a head start on "Reasoning" models (the o3 and o4 series).
The Verdict: If xAI wins, it’s because they out-built us on the industrial side—energy, satellites, and raw silicon. It’s a battle of "Software Giants" vs. a "Hardware God."
No idea what I am reading, but the divergence between Blackwell/Rubin and Google's TPUs is going to be astonishing. Can't be ignored.
Google proved the power of TPUs at least as far as Gemini is concerned and cost/energy efficiency unmatched - and only gets better with Ironwood. I’m not questioning Blackwell/Rubin from a chip perspective but seems OpenAI is paying a lot for something not needed for ChatGPT. I haven’t used ChatGPT at all since I got hooked on Gemini. I barely use Co-Pilot at this point because it feels clunky although it’s so embedded in MS I’m sure I’ll get used to it.
Watch the interview and you will see what's coming regarding NVDA vs. Google. The Blackwell version of ChatGPT is going to be a shocking step forward.
I used to own RDDT but sold when I realized I can just ask Gemini about certain topics trending on Reddit. Saves me the time of weeding through random/annoying posts and gives me a nice summary. I never have to go to Reddit meaning my eyeballs don’t see any of those ads.Those best in breed stocks are you ones you should be stacking now. CRWD, PANW, and HOOD and RDDT. No COIN.