Google, DeepMind, and the Race for AI Supremacy

Chris and Nico unpack Amazon’s new OpenAI deal, Google’s quiet strength, and how AI is already reshaping medicine, infrastructure, and software creation.

In this episode of Value First AI Daily, Chris and Nico discuss Amazon’s surprising $38 billion investment in OpenAI, the new web of cross-company alliances reshaping the AI landscape, and Google’s uniquely quiet — yet formidable — position through DeepMind.

The conversation expands into healthcare breakthroughs, energy innovation, and the accelerating democratization of software creation. In just three years since GPT-3.5 launched, humanity has learned to replicate organs, redesign materials, and vibe-code entire applications from prompts.

What You’ll Learn

1. Why Amazon’s OpenAI investment changes the balance of power in the AI ecosystem.

2. Why Google isn’t falling behind — and how DeepMind’s breakthroughs in biology, materials, and energy keep it ahead.

3. How AI is transforming healthcare, from early diagnostics to organ replication.

4. Why “vibe coding” is redefining software development, slashing cost and complexity.

5. How leaders can prepare for the next era of agile infrastructure and decentralized creation.

6. What democratized AI means for business owners, innovators, and individuals entering 2026.

Key Takeaways

• The AI race is turning into an AI alliance.

• DeepMind’s advantage isn’t market speed — it’s scientific depth.

• AI is transforming medicine from reactive to preventative.

• Hardware, healthcare, and software are converging.

• Vibe coding will redefine what “building software” means.

• The real moat in AI isn’t data — it’s agility and imagination.

In This Episode We Cover

(00:00) – Election Day intro and Amazon’s $38B OpenAI deal
(03:25) – Why rival companies like AMD, NVIDIA, and Microsoft are teaming up
(05:50) – Is Google falling behind — or quietly preparing to dominate?
(08:40) – The DeepMind factor: cracking Alzheimer’s, cancer, and new materials
(12:15) – Google’s infrastructure strategy and Project Stargate
(15:45) – How data, energy, and agility form the new AI moat
(18:50) – DeepMind’s reach into material science, fusion, and chip innovation
(22:05) – The coming reinvention of hardware as we know it
(25:10) – From protein programming to organ replication
(30:25) – AI and healthcare: from Death Clock AI to Butterfly ultrasound
(36:50) – Why the U.S. healthcare system is still structurally sick
(39:10) – How AI can make healthcare proactive instead of reactive
(42:00) – Vibe coding: the new way to create software from prompts
(47:25) – How AI-driven automation changes the economics of startups
(50:10) – Why building your own AI stack is the new entrepreneurial advantage

Transcript

Chris:

Good morning, LinkedIn friends, Value First Nation.
Welcome to another episode of Value First AI Daily, your collaborative AI intelligence report.

It’s Tuesday, November 4th, 2025—and instead of going to the polls, I guess we’re having some fun with some other stuff.
We’ll see. How are you doing, Nico?

Nico:

Doing good.

Yeah, you know—the whole vote vote thing—you gotta do it, do your duty. But for now, we’re here.

And yeah, I mean, I’m doing all right. The sphere of AI seems to be holding steady. Yet another deal from OpenAI with a very unlikely partner.

And this is the thing I’m noticing—and really like—about what’s happening with this tech sphere in particular. Unlike most corporate conglomerate deals, what I’m seeing here is real partnership and teamwork, not rivalry.

For one thing, you don’t normally see this many public deals within the same industry sector. And you certainly don’t see rivals not trying to one-up each other.

AMD and NVIDIA are both involved in OpenAI now. You’ve got Microsoft and Amazon. Amazon’s a new player—at the tune of $38 billion or so. The only one holding out at this point is Google.

I’m not sure they’ll join in—they’ve got their own knife in the back with Anthropic and everything too.

Chris:

Right.

And sorry, I don’t see your name—LinkedIn user—but I think Google may be keeping their wallet ready to buy up the best competition instead of fighting in line for OpenAI?

Nico:

Yeah, that’s more than likely Alan Parley.

And yes—LinkedIn sucks at showing names.

I could see that being the case, kind of—except DeepMind just keeps making massive breakthroughs every month. Like, every single month.

They cracked Alzheimer’s, right? Outside of cancer—which they’re also cracking—these are the biggest medical challenges on Earth.

So I don’t really know what Google is lacking, infrastructure-wise, that they’d need to invest outside their own walls.

They don’t have the same level of third-party support—they’ve been going vertical. They definitely have the money, the compute, the data centers.

When it comes to things like Project Stargate, I don’t know how much Sundar or Google are invested there. Once Stargate comes online, how much of that goes to Google compared to OpenAI?

They’re making ridiculously massive data centers. And I think what Google’s been doing, honestly, is downsizing its internal workforce to free up compute—which kind of makes sense, kind of doesn’t.

Shut down a division, you free up compute you can reallocate.

Chris:

Right—and in the world of third-party support and unified context—you’d think if they have a completed stack and infrastructure, their AI would benefit from that unified context.

Could they be planning for AI to be the third-party support layer—unless we’re talking about hardware assets?

Nico:

Exactly.

If OpenAI has NVIDIA as their top chip supplier—and AMD as second best—Google’s not a major chip company, even though they’re producing Willow and so on.

Those are built by TSMC. They don’t have full in-house hardware manufacturing. So yeah, it becomes a hardware issue—and soon a power issue.

Google’s going to have to fight for power rights like everyone else. Amazon just secured a deal in Ohio.

And as much as I love this stuff, I don’t love my power bill going up—so yes, I’m definitely going to the polls later today and attending local district meetings.

Still, I wouldn’t count them out.

DeepMind is deeper than everyone else’s. When you do the math—data + infrastructure + time in the game—they check every box.

If anything, I think Google’s keeping their wallet ready to compete, not to acquire—ready for when they launch their next big thing.

Chris:

Exactly.

This is where AI as a strategic guidance layer comes in. These companies can now model the market, predict adoption, and make fewer mistakes—unlike past cycles with VR and other premature launches.

They don’t have to guess anymore. And they’ve got friends.

Imagine if all this innovation were happening behind closed doors—they’d never get feedback from the world. Now they can let others poke the bear, learn faster.

Unless they surprise everyone by not dropping the best model of the year—which everyone expects—they’re in a strong position.

Nico:

Yeah, they’ll drop it. I don’t see how they won’t.

And the one thing people overlook? YouTube.

Owning both formats of search gives Google a ridiculous competitive advantage—massive learning data, real-time awareness, and surface area across every industry.

No one else has that.

And beyond that—they have a lead in material science. Outside of IBM, no one is as focused on core scientific breakthroughs as DeepMind.

Chris:

Right—Demis has reached “bio-digital jazz” territory.

It’s wild, but true. They’ve figured out how to program cells and DNA—going to the source of problems instead of treating symptoms.

Material science, energy, chips—DeepMind’s got investments across all of it. It’s the most agile stance imaginable.

When your moat is agility and speed, you’re ready for whatever comes next.

And at this rate, the chips we use today? We won’t even recognize them in five years.

Nico:

Exactly.

It doesn’t make sense that the tech would stay the same.

And even now, they’re going deeper—literally—manipulating DNA and protein cells at the atomic level.

They’ve figured out how to take a protein cell and replicate it.

So in the near future—maybe within two years—if you need a heart transplant, it’ll be your heart that goes back in. A replica of your heart, without the bad part.

You’ll swap out a 50-year-old part for a 20-year-old one.

And honestly—I’d sign up to get my stomach back immediately.

We’re talking about being able to repair parts of the body that were previously impossible.

Skin cancer? Practically gone.

Medical innovation is exploding.

Chris:

Right—and I just saw an ad for an app a doctor built, called Death Clock AI.

Sounds grim, but it’s fascinating. It takes your stats, compares them across massive datasets of people like you, and shows long-term health trajectories—20–30 years out.

It’s a wake-up call app.

You can literally see what your 80-year-old self looks like based on today’s choices.

And the doctor’s point made sense: healthcare costs are high because we’re treating symptoms instead of preventing illness.

Nico:

Exactly.

Health insurance rates are already ridiculous, and now they’re tripling.

From what I’ve heard, people at both the lowest and highest income levels are getting crushed.

Basically, if you’re at the stage of life when you most need healthcare—you’re about to get hit the hardest.

And that’s why I really appreciate what Congressman David Schweikert says: we’re paying so much for healthcare because this country is so sick.

The one thing we could do to lower costs is help people get healthier.

Chris:

Exactly.

And technology can do that—apps, AI diagnostics, home devices.

Bill Gates’s company built Butterfly, the first Bluetooth wireless ultrasound device.

You can scan yourself at home and send images straight to GPT for analysis.

And the results are incredible—AI has already helped people detect conditions that multiple doctors missed.

Combine these tools, and you’re getting hospital-level diagnostics from your living room.

We’re barely three years into this AI wave—and look how far we’ve come.

Nico:

Yeah—in just three years we’ve:

• Learned to replicate organs

• Discovered 700 new physical materials

• Improved power-plant efficiency by 70–80 percent

• Cut surgery recovery times by 40–60 percent

• Increased radiology accuracy dramatically

And people still say, “It’s not where we need it to be.”

Come on.

Chris:

Right—one day, though.

Nico:

Yeah—one day. Or really, today for some of us.

But yeah, give us a taste of something you’ve been thinking about building, given all these tools.

Chris:

Something I’ve been working on is personalizing software experiences—whether CRMs or anything else.

I’ve been revisiting prompt engineering—spending days planning, documenting, finding gaps.

And I realized: if you design prompts right, you can basically one-shot apps out of vibe-coders.

Users could just change variables and instantly generate the software they want—a vending-machine model for apps.

Nico:

That’s wild.

Chris:

Yeah—I’ve been prototyping it. The first one was a CRM; the second was a subscription tracker.

The CRM’s getting there—it even auto-generates records, notes, and activities. Some of the apps I was planning to build are already obsolete because the models build themselves.

For the subscription tracker, it logs in with Google, tracks real-time data, exports records, provides AI recommendations, handles Stripe payments, and manages feature access—all from a halfway decent prompt.

That’s what I’m working on now.

Nico:

And your only cost is usage—right?

Chris:

Exactly.

That’s my only cost—and it’s ridiculously low compared to traditional software development.

It used to take huge teams, hundreds of thousands in overhead, and months of work.

Now, one person can build a full stack in days.

Chris (closing):

All right—that’s where we’ll wrap today, folks.

It’s becoming clearer and clearer: if you’re starting a business, why wouldn’t you vibe-code your own stack?

More on that later this week.

Until next time—I’ll see you tomorrow morning, Wednesday, November 5th.

Have a great day, everybody.

Thanks, Nico.

Nico:

See you guys.