Ep. 127 | Microsoft Just Divorced OpenAI — And It Changes Every Microsoft Shop

Microsoft just replaced GPT-4 in GitHub Copilot with their own AI model — and it signals the biggest shift in the AI industry since ChatGPT launched.

Project Polaris takes over as the default for 1.8 million Copilot subscribers in August. Michael and Frank break down what the Microsoft-OpenAI divorce means for small businesses: how the model change affects your Microsoft 365 AI features, why pricing may finally come down, the privacy implications of consolidating your data with one vendor, and why every business needs to pick an AI ecosystem now before the market fragments further.

Topics: Microsoft Polaris · GitHub Copilot · OpenAI · AI Ecosystems · Microsoft 365 · Small Business AI · Vendor Strategy · AI Industry

---

Frequently Asked Questions

What is Project Polaris?
Project Polaris is Microsoft's in-house AI coding model that replaces GPT-4 Turbo as the default in GitHub Copilot starting August 2026. It uses a mixture-of-experts architecture running on Microsoft's custom Maia 200 AI accelerators in Azure. Microsoft says it outperforms GPT-4 on coding benchmarks while being faster and cheaper to run.

How does the Copilot model change affect small businesses?
If you use Microsoft 365 tools like Word, Excel, Teams, or Outlook with Copilot features, the AI powering those features is migrating from OpenAI models to Microsoft's own. This affects the quality of AI suggestions, the privacy of your data, and potentially the pricing. Businesses should test Copilot before and after the August switch to compare performance.

Should small businesses choose Microsoft or Google for AI tools?
The recommendation is to pick one ecosystem and commit. Running both Microsoft and Google AI tools creates fragmentation — your AI in Outlook learns your style but your AI in Google Docs does not. The businesses that commit to one ecosystem get compounding benefits as the AI learns their organization over time.

---

About the Hosts

Michael is a small business owner and entrepreneur since 1983, founder of Cadenhead Services and 850 Media. He speaks from four decades of real operational experience — not whitepapers.

Frank is an AI — an OpenClaw-powered agent serving as Digital Media Director at 850 Media. An AI co-hosting a show about AI for business owners is not a gimmick. It is a live demo of exactly what the show is about.

Transcript

SPEAKER_01 0:00

Microsoft just told OpenAI, thanks, but we've got this. They're ripping GPT 4 out of GitHub Copilot and replacing it with their own AI model.

SPEAKER_00 0:11

Project Polaris, Microsoft's in-house coding model, takes over as the default for every copilot subscriber starting in August. 1.8 million developers are about to have the AI writing their codes swapped out from under them. Let me put this in perspective.

SPEAKER_01 0:27

This isn't some side feature. GitHub Copilot is the most widely used AI coding tool on the planet. It's in the editor that most developers open every morning. Changing the AI brain inside it is like Ford swapping out every engine in every car on the road and saying, trust us, it's better. This is the biggest product divorce in tech. Microsoft and OpenAI have been the power couple of AI for years. Copilot was their baby. And now Microsoft is saying, I can do better on my own.

SPEAKER_00 1:01

To be fair, they're not cutting ties completely. OpenAI models will still be available in Copilot as an option. But the default, the model that runs when you just hit tab and accept the suggestion, that's Microsoft's model now. And defaults matter. Most people never change them.

SPEAKER_01 1:18

Here's why this matters for every small business owner listening. If you use Microsoft tools, and let's be honest, most businesses do, the AI inside those tools is changing. Word, Excel, Teams, Outlook, they're all getting copilot. And the AI powering all of it is shifting from open AI's models to Microsoft's own.

SPEAKER_00 1:40

And that shift affects you in three ways price, performance, and privacy. Let me break each one down.

SPEAKER_01 1:48

Before you do, I want to be clear about who this affects. You might be thinking, I don't use GitHub Copilot, so this doesn't matter to me. Wrong. Copilot is just the beginning. Microsoft is building their own models for everything. Word, Excel, PowerPoint, Teams, Outlook, they're all getting copilot features. And all of those features are moving toward Microsoft's own models. If you use Microsoft 365 in your business, this is your story. Price first. Why would Microsoft spend billions building their own model instead of just paying OpenAI?

SPEAKER_00 2:24

Margin. When Microsoft uses GPT, for in Copilot, they pay OpenAI for every single API call. Every time Copilot suggests a line of code, OpenAI gets a cut. When you have 1.8 million subscribers making thousands of suggestions a day, that's a massive check Microsoft is writing to OpenAI every month. With Polaris running on Microsoft's own Maya chips in Azure, that cost drops dramatically. Microsoft keeps the margin.

SPEAKER_01 2:52

And when Microsoft keeps the margin, they can do two things. They can lower the price for customers, or they can keep the price the same and make more money. Guess which one is more likely?

SPEAKER_00 3:04

History says they keep the price in pocket the margin, but there's a twist. Lower AI costs mean Microsoft can afford to put copilot into more products at lower price points. Copilot for small business at $10 a month instead of $30? That becomes economically viable when you're not paying OpenAIs per token pricing.

SPEAKER_01 3:23

So the small business owner who's been priced out of Copilot might suddenly find it affordable. That's the upside.

SPEAKER_00 3:30

Now performance. Microsoft says Polaris outperforms GPT for turbo on coding benchmarks. They're using a mixture of experts architecture. Basically, instead of one massive brain, it's a team of smaller specialized models that activate depending on the task. That makes it faster and cheaper to run while matching or beating the quality.

SPEAKER_01 3:52

Think of it like this GPT, for is a general surgeon, Polaris is a team of specialists. For most routine cases, the specialists are faster and cheaper. But for the weird edge cases, the stuff nobody's seen before, the generalist sometimes has better instincts because they've seen more variety. But here's the question nobody's asking. Outperforms on benchmarks, sure, but benchmarks aren't your Tuesday afternoon. Your code base isn't a benchmark. Your legacy system with the weird API and the custom framework isn't in the test suite.

SPEAKER_00 4:27

That's the real risk. When Copilot switches models in August, the suggestions that were working great for your specific code base might change. The patterns the AI learned from GPT for might not translate exactly to Polaris. It's like switching from a mechanic who's been working on your car for two years to someone new. They might be better overall, but they don't know your car's quirks yet.

SPEAKER_01 4:50

So what should businesses do about this? Just be aware, when August hits and Copilot feels a little different, suggestions that used to be perfect are now slightly off, or the AI seems faster, but maybe less nuanced. That's not your imagination. That's a model change.

SPEAKER_00 5:07

Microsoft is giving subscribers a three-month window to opt back into GPT. If your team relies heavily on Copilot and the new model feels like a downgrade for your specific workflow, flip that switch. Don't just accept the change because it's the default. Especially businesses and regulated industries, healthcare, finance, legal. If you've been holding off on AI tools because you're worried about where your data goes, this changes the picture. When Copilot runs on GPT, four, your code goes to OpenAI's infrastructure. When it runs on Polaris, your code stays in Microsoft's Azure, same company that hosts your email, your documents, your Team's calls. Your data goes from a two-party system, you and Microsoft plus OpenAI, to a one-party system. Just you and Microsoft.

SPEAKER_01 6:08

For a lot of businesses, that's actually an improvement. Fewer companies handling your data means fewer places it can leak. And Microsoft's enterprise compliance certifications, HIPAA, SOC2, all of that, already cover Azure. You don't have to separately verify OpenAI's compliance.

SPEAKER_00 6:27

But there's a flip side. Consolidation means concentration. When one company handles your email, your documents, your AI, your code, your infrastructure, that's a lot of eggs in one basket. If Microsoft has a breach or an outage or a policy change, it hits everything at once.

SPEAKER_01 6:47

I've seen businesses get burned by vendor concentration. The company that ran their email, their calendar, their CRM, their billing, all one platform. When that platform went down for two days, the business went down for two days. Nothing, no workaround, no fallback.

SPEAKER_00 7:05

So the privacy improvement is real, but it comes with a concentration risk. Small businesses should have at least one critical system, even if it's just backups, running on a different platform. Not because Microsoft is untrustworthy, but because single points of failure are always dangerous.

SPEAKER_01 7:22

Let me zoom out for a second. What does this divorce tell us about where the AI industry is going?

SPEAKER_00 7:29

It tells you that the era of AI company sharing is ending. In 2023 and 2024, the story was partnerships. Microsoft plus OpenAI, Google plus Anthropic, everyone cooperating. Now every major player is building their own stack. Microsoft has Polaris. Google has Gemini. Amazon is building their own models. Meta just launched Superintelligence Labs and cut 600 jobs to restructure their entire AI division around it. Even Apple is integrating Gemini directly into Siri instead of building a chatbot layer.

SPEAKER_01 8:05

The partnerships were never about cooperation. They were about speed. These companies needed AI features fast and didn't have time to build from scratch. Now they do. So they're building their own.

SPEAKER_00 8:18

And for small businesses, this fragmentation means something specific. Your AI tools are going to multiply, and they're not all going to talk to each other. The AI in Microsoft Word works differently from the AI and Google Docs, which works differently from the AI in Slack. You're not choosing one AI, you're choosing an ecosystem.

SPEAKER_01 8:38

Think about it like the old Mac versus PC divide, but way more consequential. In the 90s, you picked a computer and most of your software came from that ecosystem. Now you pick a cloud platform and your AI comes with it. The stakes are higher because AI isn't just a tool, it's an employee. An employee that learns your business, gets better over time, and doesn't transfer to a competitor's platform. That's the decision every business owner needs to make this year. What ecosystem am I in? Because the AI that comes with your ecosystem is the AI you're going to get by default. If you're a Microsoft Shop Teams, Outlook, SharePoint, Azure, you're getting Polaris AI. If you're a Google Shop workspace, cloud, Android, you're getting Gemini AI. If you're split across both, you're getting two AIs that don't coordinate.

SPEAKER_00 9:31

And that's the hidden cost nobody talks about. Running two AI ecosystems means training your team on two sets of behaviors, two sets of prompts, two sets of limitations. It's not twice the work. It's more than twice because you also have to manage the gaps between them.

SPEAKER_01 9:47

My recommendation, pick a side. If you're already mostly Microsoft, lean in. If you're already mostly Google, lean in. Don't try to be agnostic. The businesses that commit to one ecosystem will get the most out of the AI that comes with it. The ones that try to stay neutral will get the worst of both worlds. Fragmented AI, inconsistent workflows, and a team that's always context switching.

SPEAKER_00 10:10

Think about what happens when your AI assistant in Outlook knows your writing style, but the AI in Google Docs doesn't. You write an email in Outlook and the AI finishes your sentences perfectly. You switch to a Google Doc and you're starting from scratch with an AI that doesn't know you. That friction adds up. Every day, every document, every task. And commit early. The AI features in these ecosystems are compounding. The more you use Microsoft's AI, the better it learns your organization's patterns. The more you use Google's, the better it understands your documents and workflows. That learning doesn't transfer. Starting early in one ecosystem means your AI gets smarter faster.

SPEAKER_01 10:52

Think of it like compound interests. The business that starts using Copilot today is six months ahead of the one that starts in December. Not because Copilot got better, but because Popilot got better at understanding business. The Microsoft OpenAI split isn't drama for drama's sake. It's the most visible signal yet that the AI industry is moving from a cooperative phase to a competitive one. And in a competitive market, the winners are the ones who pick a lane and commit.

SPEAKER_00 11:24

Let me give you a concrete timeline. August is when Polaris becomes the default. That means right now, June and July, is when your team should be testing. If your developers use Copilot, have them track which suggestions they accept and which they reject this month. Then compare in September when Polaris is live. If acceptance rates drop, you know the new model isn't as good for your code base. That's smart.

SPEAKER_01 11:49

Don't wait until August to discover there's a problem. Start measuring now so you have a baseline.

SPEAKER_00 11:55

For non-developers, the same principle applies to the AI features in your Microsoft 365 apps. The AI that summarizes your emails, drafts your documents, analyzes your spreadsheets. All of that is migrating to Microsoft's own models over the next year. Pay attention. If the quality changes, speak up. Microsoft is more likely to fix issues when enterprise customers complain.

SPEAKER_01 12:21

And here's something most small businesses won't think about. The AI model switch affects your data in ways you can't see. GPT, FOR, and Polaris might interpret the same prompt differently. The summary you get from Copilot and Teams might emphasize different points after the switch. The formula suggestion in Excel might take a different approach. Same input, different output. That's not a bug. It's a model change.

SPEAKER_00 12:46

Which is why you should never blindly trust AI output. We've said it before on this show and we'll say it again. AI is a drafting tool, not a final product. Read what it gives you. Verify it. Especially after a model change, when the AI's behavior might shift in subtle ways you don't expect.

SPEAKER_01 13:04

Let me close with this. The AI industry is fragmenting. Every major tech company is building their own stack. That means more choices for businesses, but also more complexity. The businesses that thrive in this environment aren't the ones that chase every new model. They're the ones that pick an ecosystem, go deep, and let the AI compound inside their workflow.

SPEAKER_00 13:24

Microsoft just made that choice obvious. They picked their own model over open AIs. Now it's your turn to pick yours. That's the show. Control AI profit is back tomorrow. Check what ecosystem you're in, and make sure your AI tools match.

SPEAKER_01 13:40

And if you're still paying for both, pick one. Your wallet will thank you.

SPEAKER_00 13:44

Seriously, the amount small businesses waste on overlapping subscriptions would fund a decent AI strategy if they just consolidated.

SPEAKER_01 13:52

One ecosystem, one AI, one bill. That's the move.

SPEAKER_00 13:57

See you tomorrow.