Streamline & Scale Newsletter: Issue 12: Podcast

So right now, pretty much every single company in your sector is buying AI tools.

Oh, absolutely.

I mean, they're trying out ChatGPT, they're paying for smart email assistants, plugging chatbots into their websites. But, we've got some new data from Reuters Insights, and it says only 6% of those companies are actually gaining a competitive structural advantage from all of this.

Right. Just 6%.

Yeah, and the other 94%, they're just kinda treading water. So today, we're finding out why. We're diving into Red Hawk Technologies monthly newsletter, Streamline and Scale, Volume two, Issue 12.

Which is a great issue, by the way.

It really is. Wonderful. And alongside the newsletter, we're also pulling from a supplementary piece by Ryan Frederick.

That one's called Beyond Tools: Why Your Business Needs an AI Operating System.

Yeah, that article really brings it all together.

Exactly. So our mission for this deep dive is to figure out what that top 6% is doing differently, right?

Yeah.

And, how you can stop just playing around with AI tools and start fundamentally rewiring your [00:01:00] business.

Yeah, because that 6% statistic, I mean, that is the wake-up call the market desperately needs right now.

For sure.

We're officially moving past the whole novelty phase of artificial intelligence. If a business leader's strategy is simply telling their team to just find an AI tool to save some time, well, they're already falling behind.

Right, Okay, let's unpack this. The newsletter calls this the trap of tool adoption.

Right, the tool trap.

Yeah, where companies are just buying these standalone applications to solve specific, isolated problems.

Yeah.

But I'll be honest, if I'm a mid-market CEO and my sales team starts using a tool to summarize hour-long discovery calls into, I don't know, a one-page brief in 10 seconds-

Oh

that feels like a massive win.

Oh, yeah. It definitely feels like one.

Right. My team gets faster. We save hours of busy work. Productivity goes up. So why are the sources calling that a trap? I mean, why fix what isn't broken?

Well, the gains are definitely real, and that's exactly what makes it a trap. It gives you this false sense of security.

Ah, I see.

Yeah, [00:02:00] because when you buy a standalone tool to, say, summarize PDFs or draft emails You're engaging in what economists call consumption. Yeah. You buy a capability, you use it, and you get a fixed one-time benefit.

Right. It's just a transaction.

Exactly.

Right.

Yes, your employee saved an hour today, but the tool itself doesn't get better tomorrow.

It doesn't, you know, learn about the weird nuances of your specific industry over the next six months.

Yeah, that makes sense.

And crucially, it doesn't create any kind of structural advantage that your competitors can't just replicate by buying the exact same twenty dollar subscription tomorrow.

Wow.

Yeah. It's kinda like hiring a brilliant temporary contractor, right? They come in, they do a great job drafting your emails, and then they leave at 5:00 PM every day.

Right. They just walk out the door.

And they take their brain with them. The next morning, you have to prompt them all over again. They never actually learn the fabric of your company.

Mm-hmm

and you're just renting their intelligence day by day. You aren't actually building anything permanent.

That is the [00:03:00] perfect way to look at it. You're just renting isolated moments of productivity, and that connects directly to the core economic argument in the newsletter, which is about how a business scales.

Yeah, you're right.

Because traditionally, scaling is very linear. If a company wants to double its output, well, it generally has to double its headcount, right? Which means a proportional doubling of operational costs.

Yeah, it's a one-to-one ratio.

Right. And even if those employees are using smart tools, your output is still fundamentally capped by the physical limits of your human capital.

Which is what the ninety-four percent are doing.

Exactly. What's fascinating here is that forms of power that used to compound things like sheer headcount or mature internal processes or just decades of institutional scale, they just don't compound the way they used to.

Wait, really?

Yeah. Those things are valuable, but they don't give you exponential growth anymore.

The top six percent of companies are moving away from that linear scaling model. They're moving toward compounding intelligence.

Mm-hmm. Compounding. I mean, we hear that in [00:04:00] finance all the time, obviously, but applying it to organizational intelligence is a little, uh, hard to wrap my head around. How does a company's capability actually compound?

Well, compounding intelligence means you're embedding AI directly into the operating model of the business itself rather than just handing it to employees as a little add-on accessory.

Okay.

When it's structurally embedded, the system actually learns from every single transaction, every customer interaction, every mistake.

The entire organization gets smarter every single day, and you aren't adding a proportional cost in human labor.

Okay, so if standalone tools are the temporary contractors in this analogy, what does this compounding system actually look like? The newsletter and Ryan Frederick's article both point to this specific solution, right?

Yeah. An AI operating system or AI OS, which uses a framework called TransformOS.

Yes. TransformOS is the framework.

I mean, for a listener hearing operating system, that sounds incredibly abstract. Sounds like right [00:05:00] now companies are just screwing in a bunch of smart light bulbs into old sockets, but what they actually need to do is completely rewire the house.

That is a great analogy. You're literally rewiring the house.

Right. So break that down for us. If we aren't just buying tools- Mm. -what are we mechanically building here? What makes up this new wiring in the TransformOS framework?

Well, we have to stop thinking of AI as just software you log into. It's more like an invisible layer of connective tissue running beneath the whole business, and the text outlines six interconnected components to actually build that layer.

Okay, six components. Let's go through them.

So first, you have the models. These are the reasoning engines, like the large language models.

The brains, basically.

Right. But a brain can't do anything without hands, which brings us to the second layer: agents. These are autonomous workflows acting without human direction.

Ah, okay. So the agents do the actual work.

Exactly. Then the third component is proprietary data. This is your organization's [00:06:00] structured knowledge.

The stuff that makes your business unique.

Yes, your specific history and context. Without this, the AI is just guessing. Then fourth, we have redesigned workflows.

You can't just slap AI on an old process. You have to rebuild the workflow completely around the AI's capabilities.

That sounds like a pretty heavy lift.

It's an adjustment, for sure, which is why the fifth component is so critical: by judgment and governance policies.

Okay, like guardrails.

Exactly, rules that make all the AI's decisions completely auditable.

You need to know why it did what it did. And finally, the sixth layer is compute and energy access. You need infrastructure that can actually scale to support all of this.

Wow. Okay. So models, agents, proprietary data, redesigned workflows, governance, and compute.

Right. And if you remove even one of these layers, you instantly devolve back into just having a collection of disconnected tools.

It all falls apart.

Pretty much. But together, they create true organizational intelligence. The whole company gets smarter over time, [00:07:00] not just the individual employees using an app.

Here's where it gets really interesting, though, because moving from that theoretical framework to real-world application Uh, the text gives a phenomenal proof of concept for this in the healthcare sector.

Oh, the prior authorization example. Yes.

Yeah. So they built a single agent to handle prior authorization paperwork, and it cut a seventy-two-hour process down to just four hours.

Which is just staggering when you think about it.

It is. But look, whenever we hear about automation capital and these massive time savings, people immediately panic about job losses.

Of course.

I mean, is this just a fancy framework for firing people? I know a lot of people listening are probably wondering that.

It's the most common reaction, but it fundamentally misunderstands the goal of an AI operating system.

Okay. How so?

Well, the specific detail in the text addresses this head-on.

The staff who were doing all that manual paperwork, they didn't lose their jobs. They were actually promoted to case supervisors.

Oh, wow. So they stopped being like data entry typists [00:08:00] and became exception handlers.

Exactly. The system handles the massive, boring volume of standard authorizations, and the humans handle the edge cases, the really complex situations that actually require human empathy and ethical judgment.

Okay, that makes sense.

And this is how proprietary data and feedback loops actually work. The system handles the bulk, the humans handle the exceptions, and the system learns from the humans.

Wait, so when the human supervisor makes a decision on a weird edge case- That action gets recorded.

Right. It immediately updates the proprietary data layer.

The system observes the human's logic, digests the new parameters, and adapts.

So the next time that exact same weird edge case comes up, the agent already knows what to do.

Exactly. It continually improves without you having to add proportional headcount cost.

Man, that is incredibly compelling. But, it also feels like the kind of technology reserved for Fortune 500 companies with, you know, massive IT budgets.

Mid-market companies usually have to wait years for this stuff.

[00:09:00] Traditionally, yes.

Which brings us to the massive strategic news in issue twelve. Red Hawk Technologies has officially acquired Transform Labs, the exact team that built the TransformOS framework we've been talking about.

It's a huge move.

Yeah. I read Matt's welcome letter in the newsletter, and he explains that this acquisition allows them to bring enterprise-grade AI to their mid-market clients right now, not five years down the road.

Right. They're commoditizing the architecture required to build that structural advantage.

But okay, let's ask a really practical question for the listener.

Say I'm a mid-market operations director. I've got a massive, clunky legacy ERP system and a CRM that my team barely uses.

The usual setup, yeah.

Right. Does a mid-market company have to rip out their entire legacy tech stack to build this AI OS? Because that sounds like a nightmare.

No, and that is a crucial point.

You do not have to rip and replace your existing infrastructure.

Oh, thank goodness.

Right. TransformOS is designed specifically to layer [00:10:00] onto and integrate with your existing tech stack. The old tools can stay. What changes is how they're integrated.

Okay, so how do you even start then? If you aren't ripping it all out, what's step one?

The newsletter's advice here is very clear. Don't boil the ocean.

Right.

Don't try to transform the whole company on day one. Identify just one single high-leverage workflow, like that prior authorization process or maybe vendor invoicing.

Just one process.

Yes. Redesign that one workflow using the framework and build from there.

Get the quick win, prove the concept, and then expand.

That makes it so much more approachable. And speaking of approachability, I wanna make sure we highlight the upcoming Dayton AI Day on August 26, 2026.

Yes, that's gonna be a great event.

Yeah. Matt is actually gonna be speaking there about Red Hawk's own raw real-world journey.

The session is about moving from manual spreadsheets to a custom ERP.

Right. It's not just a theoretical pitch.

Exactly. It's a behind-the-scenes look at the [00:11:00] friction and the cultural hurdles of actually changing how work gets done. So if you're feeling overwhelmed that session is gonna be incredibly grounding.

I completely agree. Seeing someone else's messy transition really validates the friction every company faces.

So what does this all mean? Let's bring it all together for the listener.

Sure.

Adopting standalone tools gives you measurable progress, right? It makes employees faster.

Mm.

But building an AI operating system gives you leverage.

It shifts a business from just having faster employees to becoming a fundamentally smarter organization overall.

That's the core of the paradigm shift, and it really leaves you with a stark mathematical reality to mull over.

Oh, yeah. Let's hear it.

Well, if an AI operating system literally learns and improves with every single decision it makes, what happens to the gap between companies?

Oh, man.

If you wait three years to stop buying tools and start building a system, will it even be mathematically possible to catch up to a competitor whose [00:12:00] structural intelligence has been compounding that whole time?

Wow. Yeah, if they're getting exponentially smarter every day, and you just have a slightly faster email drafter next year, that's an uncrossable gulf.

Exactly. The math gets terrifying very quickly.

Well, on that slightly terrifying but very motivating note, thank you for joining us for this deep dive. We highly encourage you to check out the full volume two, issue twelve of Streamline and Scale to explore the TransformOS Playbook for yourself. Because while everyone else is renting temporary intelligence, it is definitely time for you to start rewiring the house.