October 28, 2026
Bonus Content: Nvidia Just Got a New Partner. And a New Rival.
Every American Should Unlock Their IRA/401k In 2026
Most Americans have no idea their Retirement Plan is locked down.
Let’s say you’re at a restaurant – you can order anything you want, as long as it’s on the menu.
But what if you found out you have a fish allergy, and you’re at a seafood restaurant?
Because if you’ve had an employer 401(k), then you remember seeing the options:
You can invest in anything you want, as long as it’s on the menu. Same thing with many IRA plans.
But what happens if everything on the menu can make you sick?
That’s what many Americans are facing right now. Because whether you order the S&P Index Fund, the 60/40 Stocks and Bonds mix, or any other combination on the menu…
All the ingredients come from the same kitchen – the stock market.
And you already know the stock market has made people sick.
The ups and downs can be nauseating.
Now that’s not a problem if you’re a few decades away from retirement, because you can ride out the storm, and things could likely turn in your favor.
But when you’re already retired, or about to retire, then the game changes.
Because it’s not about growing your assets anymore, it’s about preserving your assets, right?
That means if you have $1 million saved for retirement…
Then you expect that million to pay for your travel, gifts, visits to family and grandkids…
For the rest of your life. Literally! And that’s a great plan.
So how do you tackle the problem?
You could hope the stock market rockets up again, and grows your retirement…
Or you could look at assets that help defend against inflation.
Except the problem is, many IRAs and 401(k)s can be locked out of inflation-defending assets.
But it doesn’t have to be that way.
Because right now, thousands of Americans have figured out how to UNLOCK their IRA/401(k) so they can diversify a portion of their retirement savings into assets that help defend against inflation.
That’s why we put together this free guide.
It’s a brief, to-the-point handbook that shows you:
- The IRS Rule That Lets You Access Locked Retirement Funds
- Who Really Controls Your Retirement Savings
- How To Transition To An Unlocked IRA
And it’s completely free to get a copy.
So what you do with your retirement savings is your business, but at the very least, you should be educated about your options.
But of course, the choice is yours.
Nvidia Just Got a New Partner. And a New Rival.
On Tuesday in Las Vegas, Oracle held a single event that did two things at once: it deepened its relationship with Nvidia and invited Google’s silicon onto the same platform. Fund managers who have spent 2026 debating whether a credible alternative training chip changes Nvidia’s valuation story now have a concrete test case to work with.
Why This Matters to Institutional Investors
Oracle and Google announced that Google’s Tensor Processing Units will be available through Oracle Cloud Infrastructure, marking the first time Google TPUs will be offered on a hyperscale cloud platform outside Google. That is not a minor footnote. For three years, accessing a TPU meant signing into Google Cloud. Google TPUs on OCI will be available in calendar year 2027, ramping over a multi-year period to meet customer needs. The addressable market just widened.
On the same day, Nvidia and Oracle announced plans to expand Oracle Cloud Infrastructure Superclusters with Nvidia Vera Rubin NVL72 systems to multi-gigawatt scale, supporting growing customer demand for AI training, inference, and agentic applications. Oracle and Nvidia also expanded their AI infrastructure collaboration with the general availability of Nvidia Vera Rubin NVL72 and Nvidia Vera CPUs on OCI.
The Bull Case for Nvidia
The multi-gigawatt Vera Rubin commitment is not a promise, it is a buildout plan. Oracle and Nvidia are clear about the ambition, but neither company provided a timetable, site list, or current installed gigawatt figure tied specifically to Vera Rubin. Oracle said Vera Rubin delivers up to 30 times higher AI-factory throughput per megawatt than Nvidia GB300 NVL72 preview results, and it has highlighted large training-performance gains versus prior generations. That generational gap is hard for any rival to close on a one-year roadmap. The Vera Rubin expansion builds on OCI’s existing Nvidia infrastructure, and Nvidia’s CUDA software foundation provides continuity across hardware generations. Customers do not rewrite model code when they upgrade racks. That software lock matters more than any single benchmark.
Nvidia reported $193.7 billion in revenue for its fiscal year that closed in January 2026. But the claim that Nvidia controls roughly 81% of the data center AI chip market is not something the companies involved report in a consistent, auditable way, and it depends on definitions (GPUs versus accelerators, training versus inference, and revenue versus units). Jensen Huang has argued repeatedly that the Nvidia stack offers the best performance-per-total cost of ownership ratio on the market. The Oracle commitment to gigawatt-class Vera Rubin capacity reinforces that position at exactly the moment a rival chip is arriving on the same platform.
The revenue figure alone does not settle the valuation debate. Nvidia’s gross-margin guide has already shown how a single CFO line can complicate an otherwise clean earnings sweep.
The Bear Case
The concern is not that TPUs will win. It is that the conversation changes. Google’s eighth-generation TPU includes two architectures: TPU 8t, designed for training workloads, and TPU 8i, designed for inference. Google has positioned TPU 8t as a way to shorten development cycles for frontier models, while TPU 8i is designed for inference workloads in the agentic era. Inference is where pricing pressure on Nvidia is most acute.
The announcement did not specify an initial TPU model, launch quarter, chip count, pricing, customer orders, or expected revenue. The deliberate vagueness makes it impossible to size the threat today. But the structural shift is real: companies building AI systems could now run different chip architectures inside a single cloud platform, rather than stitching together multiple providers. That lowers the switching cost between Nvidia and its rivals.
The switching-cost question extends beyond Oracle’s platform. Amazon’s in-house Trainium and Graviton lines represent a parallel bet that cloud providers can capture chip economics without depending on a single silicon vendor.
What Investors Are Missing
The debate has been framed as Nvidia versus TPUs. The more interesting question is what Oracle gains from hosting both. Using the roughly $40 billion per gigawatt revenue opportunity Nvidia cited on its August 26, 2026 earnings call as a benchmark, a single gigawatt of Vera Rubin would be a meaningful slice of Oracle’s fiscal 2027 capital spending plan. Oracle does not need to pick a winner in the chip war; it needs utilization. Offering enterprise customers a genuine choice of silicon on a single platform with shared networking, billing, and security is a procurement argument Oracle’s rivals cannot easily match today.
Stocks to Watch
- Nvidia (NVDA): Multi-gigawatt OCI commitment confirms demand at scale. The real question is whether inference pricing compresses as TPU 8i reaches enterprise customers in 2027.
- Oracle (ORCL): Positioned as the neutral ground in the chip war. If customers want flexibility across Nvidia and Google silicon without juggling two clouds, OCI is making that case.
- Alphabet/Google (GOOGL): OCI distribution expands the TPU addressable market beyond Google Cloud for the first time. Revenue upside is unquantified but the strategic signal is significant: Google is treating its silicon as a product, not a moat.
