AI Made Easy for Business

1. SUMMARY
In this episode, Jeremy and Jim dive into the tension between AI hype, mass delusion, and grounded innovation. With Jeff Bezos calling AI an “industrial bubble,” and The Atlantic warning of cultural anxiety, the hosts explore whether this boom is all smoke—or fuel for something more lasting. They break down Google’s new Gemini Enterprise suite for small businesses, revealing how to use it (or any LLM) for real tasks like sales forecasting and inventory planning. Plus: how to filter AI news without getting lost in the noise, and why predictable costs and automation-ready workflows are a win for SMBs.

2. THE AI JOKE
What do you call a Gemini who's also a DJ?
 A remix.

3. TAKEAWAYS
  • Jeff Bezos frames AI as a bubble, but one likely to drive long-term innovation—not collapse.
  • The Atlantic's “mass delusion” thesis calls out AI’s role in creating digital illusions and cultural anxiety.
  • Google’s Gemini Enterprise bundles AI tools for businesses at $21–$30/user, aiming to simplify AI adoption.
  • Predictable per-seat pricing can help SMBs budget for AI without surprises.
  • “Seat creep” is real—scale wisely and measure ROI before rolling out to larger teams.
  • Tools like Gemini offer no/low-code agent-building options to automate summaries, routing, and data tasks.
  • Beware of fake AI apps—especially post-launch. Only download from trusted sources.
  • The “Sales Pulseboard” tip shows how to merge POS data with local context to make smarter staffing/inventory calls using LLMs.
4. RESOURCES AND TOOLS
5. CHAPTERS
00:00 – AI Joke: Remix and Gemini Wordplay
00:23 – Bezos on AI: Bubble or Breakthrough?
01:10 – Survivors of the Bubble: Risks and Optimism
02:12 – AI Hype and Cultural Anxiety (The Atlantic)
03:21 – Filtering the BS for Small Businesses
04:28 – Real Talk: How AI Investment Affects SMBs
05:30 – Hallucinations and the Case for Workflow Checks
06:49 – App Store Fakes and AI Safety Tips
07:23 – Gemini Enterprise Explained: Pricing and Bundles
09:18 – Seat Creep: Watch Those Subscription Costs
10:30 – Predictable Costs & Free AI Training for SMBs
11:52 – Tool Tip: Building a Sales Pulseboard
13:38 – External Factors: Events, Weather, and Demand
14:59 – Custom GPTs and Automation Possibilities
16:25 – Why It’s Worth It: Long-Term Automation Wins

6. TOOLTIP
Quick Win: “Sales Pulseboard”
What it is:
A one-page Google Sheet that merges your last 30–60 days of sales with simple local context (events + weather) and a prompt you can paste into an LLM (Gemini/Claude/ChatGPT) to get plain-English staffing and inventory suggestions.
Why it helps:
Catches demand spikes (market days, local games, heat waves) so you can tweak staffing, hours, or SKUs—without buying new software.
You’ll need (free or freemium):
  • Google Sheets
  • Any LLM in your browser (Gemini, Claude, or ChatGPT)
  • Optional: links to local event calendars + a 7-day forecast page
Steps:
  1. Export sales from your POS as CSV for the last 30–60 days. In Sheets, add columns: Date, Hour, Category, Item, Qty, Net Sales.
  2. Summarize with a Pivot Table:
    • Rows = Date
    • Columns = Category (or Hour)
    • Values = Sum of Net Sales and Sum of Qty
  3. Add context tabs:
    • Events: Paste dates/times of nearby events (games, festivals, farmer’s markets).
    • Weather: Paste 7-day highs/lows + notes (rain, heat, storms).
  4. Prompt the LLM (paste this, or upload the CSV/Sheet link if supported):
    You are a retail operations analyst. Using the “Sales”, “Events”, and “Weather” tabs, find day-of-week and hour patterns, flag correlations with events/weather, forecast the next 2 weeks at a high level, and output 5–8 specific actions (staffing, hours, inventory, promos). Include a simple table with predicted busy periods and a confidence score for each.
  5. Act on 2–3 changes this week (e.g., add a Friday evening shift, pre-prep top SKUs) and set a calendar reminder to review results next week.
  6. Streamline: Once you've perfected the prompt and steps after a few weeks, make a custom GPT or Gemini app, so that all you'll need to do each week is export the sales data, upload to the GPT, and go!
  7. Optional automation: Paste the AI’s recommendations into a one-page doc and share to your team Slack/WhatsApp every Monday.

Creators and Guests

Host
Jeremy Ryan
CIO at Polarity, having fun with innovation, creative technology, and AI (through hard word)
Host
Jim Donio
CEO at Polarity, serial entrepreneur, equity partner, and award-winning leader in business innovation

What is AI Made Easy for Business?

”AI Made Easy for Small Business” by Polarity.biz is a lively podcast hosted by successful small business owners, Jim Donio and Jeremy Ryan, offering practical insights and strategies to help fellow entrepreneurs harness the power of AI, boost competitiveness, and drive growth—all with a dash of humor and fun. No prior AI experience is necessary.

Jeremy (00:00)
What do you call a Gemini who's also a DJ? A remix. You don't get... Gemini is a play on words, right? Gemini the technology, Gemini twinning. ⁓

Jim (00:02)
I don't know what.

Get it? Well, no, feel.

But that's more my own.

Jeremy (00:15)
Bold and dead

inside. ⁓

Jim (00:23)
All right, so Bezos is weighed in. If Bezos from Amazon is weighed in. But I mean, I did. actually, maybe, maybe, maybe, yeah, on AI, he said, we're, you know, it's maybe in an industrial bubble, which kind of.

Jeremy (00:28)
But is anybody listening? ⁓

He's not the only one to be saying this. There's a lot of people who saying this, you know, there's a lot of people who said that the internet wouldn't be a thing either. There was, there was a dot-bomb.

Jim (00:44)
There was a bubble. He made it to the

other side of it, though. Right. You think about it with Amazon. But through just hundreds and millions and billions and billions of dollars of backing to make it happen. So Financial Times just basically said that, you know, the investment in AI is definitely almost definitely going to be a bubble. Like someone's going to get left holding the bag. But that doesn't mean that there won't be survivors and companies and people that make it to the other side. There's going to be a lot of losers.

Jeremy (01:10)
But it's also

the ⁓ what's his face? ⁓ Cut open AI. Thank you. He also said ⁓ more millionaires we've made in this than any other technology before it. So, you know, I. I hope. Right.

Jim (01:25)
That's plausible. I million. I used to be. All

right. So that's not really I mean, that was just kind of a throw away one, but like one that we thought I saw and I thought, you know, when when again, the more people that are recognizable that weigh in and cut through the clutter and also acknowledge like, hey, it's probably going to blow up. But also, like, it's still going to be transformational. That's pretty much what he said is, no, it's transformational. There's going to be ups and downs. And I think that trajectory I had a conversation last night about this. It's a trajectory with AI won't

be like this, it's going to be a little windy and I actually think it is going to be like this but I do think there will be some big sharp drop.

Jeremy (02:03)
I

just like your dance moves. For those listening in, Jim is doing the one man wave. worm. The one armed worm.

Jim (02:12)
And then the other the other other are going to we're not going to focus on today is ⁓ a piece in the Atlantic about ⁓ AI fueling cultural anxiety. And I like this one for a lot of reasons, because I do think that it is weighing. Yeah, they're all unrelated, just Bezos in general is cultural anxiety. It's a ⁓ no, just the mass delusion ⁓ that that, you know. That it's just a social force that is this is this.

Jeremy (02:23)
Well, isn't one related to two?

Jim (02:39)
ever present, all knowing and just all powerful. And it's just going to overtake everything or it's going to overtake everything in a really negative way or a positive way. And just. That it creates a lot of illusion, right? People literally creates. Literally in digital language, it's creating illusions, right? Absolutely nations, right? We see that we're just making up fake responses or fake citations and things like that.

Jeremy (03:05)
Well, Sora 2 promises to be an entire platform based on illusions and...

Jim (03:11)
So

I don't know. thought it was kind of but it's just like it was a warning but also like yeah fair but also it's not gonna go away I mean, what do think the regulators are gonna do anything? Exactly and I think again you're up a way in but basically in the US largely it's gonna be left unchecked

Jeremy (03:21)
Well, not these regulators.

You know, I think the important takeaway here for small businesses as we're thinking about both of these items are number one, ⁓ when Bezos or Bezos or whatever, when that dude is, I don't know him, so don't, know, whatever. ⁓ Though I shop on Amazon and I use AWS. But yeah, but this idea that

Jim (03:44)
Kind of well known.

as everyone else.

Jeremy (03:58)
he's thinking about the ups and downs. I think it's important for our listeners who are likely small businesses like us that he's talking about enterprise. And when we think about these ups and downs. he's also talking about the investing that happens at large scale. Corporation, corporate investments.

Jim (04:11)
big corporations.

So, but also. Absolutely.

Corporate investments in this technology or backing this technology or implementing this technology.

Jeremy (04:28)
And

so when we're thinking about the impact to small business, I would say one of the other important discerning characteristics that is necessary in the AI age isn't just about being critical for your own purposes or for your business's purposes when thinking about technology investment, but also filtering out a lot of the BS in the news sphere that isn't applicable to small business. I love these articles because I think we do have a national pulse at this

regardless of the size of your business. And we all participate in that pulse. I also think it's important to sort of cut out the chatter. The article that we didn't get to this week, I didn't plan on talking about this because we left it out. But just to sort of mention this, and maybe we will talk about it in a future episode, is OpenAI's new agent kit. ⁓

And the reason why I'm mentioning it is there is a built-in hallucination check that you can build into your workflow. It's not just nice, it's also kind of like it's its own recognition that this is a BS engine. It's a fantastic and worthwhile piece of technology, but it has the propensity to go off the road.

Jim (05:30)
Which is nice.

How

much BS do we think is in a lot of work product right now anyway, I mean, let's be perfectly frank, right? Like what is the likelihood that? work product like this would be ⁓ Wouldn't be by humans bullshit inserted into it to abs. No, I when we say sometimes Hallucinations like right, but a lot of people just make shit up into what step or product to? Wing it and Google it and copy and paste from whole reliable sources all the time

Jeremy (06:00)
absolutely.

I would go with. I would go with.

Well,

one of the other things that I was reflecting on as we were thinking about maybe doing ⁓ a segment on Sora 2 is I was investigating. I'm an Android user, as you know, and Sora 2 is not yet available for ⁓ native app on Android ⁓ to my chagrin, because I'd like to play with the technology at least. ⁓ But what I found looking at the app store is the

ridiculous proliferation of fake Sora apps that immediately happen right on the heels of some kind of this type of announcement. And it's not just this announcement, it's all.

Jim (06:49)
or something pops. I mean, listen, chat CPT happens all the time. I always tell people, hey, if you haven't downloaded chat CPT or any of the tools that we might recommend, I'm like, just pause and make sure I've had friends in a business associate to download the wrong one. they're like, yeah, mine doesn't look like yours. I'm like, OK, yeah, but you downloaded the fake one. You do need to learn.

Jeremy (07:08)
Yeah, and so my point is just I do think that humans will create these fake things whether they have AI to help them with it or not.

Jim (07:23)
So what we're mainly gonna be talking about today is this article about Gemini Enterprise. It's in TechCrunch, but basically Gemini Enterprise is Google and what they've done to bundle for businesses, enterprise for small and medium sized businesses. Enterprise is just business, right, but usually bigger business. But in this case it's meant just business as a whole. And the idea is that in this article what's covered is that Google is really making a play to continue to build out an ecosystem.

whether it's masked with your domain or not, and you're probably using Drive or Docs and different things, you know, Calendar, but this gives you an opportunity to now bundle some AI tools all in one, like Gemini and other things as well, right? ⁓ And so...

It seems like even for larger enterprises, larger companies like Gap are using it in other companies. But as a small business owner, this is a great way for to jump in and do it in an easier way at a relatively affordable price. What's the price?

Jeremy (08:23)
Well, like the plus tools are 30 bucks a month, but even basic bundled Gemini is like 21 bucks per user per month. And I think that's really the whole point of this is ⁓ if you're paying for multiple LLMs like ⁓ OpenAI's ChatGBT, and if you're using Microsoft Office, for example, Google wants your business because they're gonna give you Gemini and their version of Office, right? Google Workspace, now Gemini Workspace.

I'll it to you as one big tool. Somebody, ⁓ a business like ours, we're already on Google's ⁓ ecosystem for Google Workspace. So for us, it's a nice little upgrade, but yes, we're paying more per month because it's 16 bucks per head on Google Workspace, now 21 bucks with Gemini or 30 bucks for the Plus model. We're gonna get seat creep, I mean, which I love the name.

Jim (09:18)
This what

this is. Our podbot use our little art to create the content but that helps us all along the way. And the idea is that basically, right, idea of SeatCreep is basically just the idea that, it could quickly add up.

Jeremy (09:21)
Beware of sea creep. I love the double entendre.

Soloprenors

are going to look at 30 bucks and go, OK, not a big deal. But if you have like 10, 20, 50 employees.

Jim (09:40)
Pretty fast. Yeah, but again if you're already using it for their email accounts and for their drive and meat and other things then it kind of it's not really that bad and if you don't you let them use chat CPT and you focus some more on Gemini tools and Google's AI then now it's kind of not that terrible

Jeremy (09:56)
But if you're using Office 365 and you're paying for that subscription and then you're also paying for either Claude, my personal favorite right now, or ChatGP, right, all these things bundled together, they actually will help you reduce the amount of cost and the seat creep notion actually. Right, that's the whole point is consolidation.

Jim (10:12)
those other things. Creates

an ecosystem in one under Google or Alphabet or whatever they're calling themselves. Rebranding is now Gemini for this purpose. But the point is, the idea is to really make it easier for small, medium sized businesses to start to adopt and adapt to AI.

Jeremy (10:30)
And I also think one of the other components of this is predictable cost, right? So yes, maybe you're not used to paying 30 bucks a month per seat ⁓ per user for this, but ⁓ if you know that that's what you're really going to pay and that's the extent of it, if your costs are unpredictable, having this predictable cost model actually might help you. The other nice thing about it is, and it's just worth noting, I'll put it in the notes, ⁓ that Google is offering ⁓ SMB, small and medium sized businesses,

this AI training. They've earmarked $5 million for 40,000 US businesses and called.

Jim (11:07)
It sounds like a lot of money when you really think about how much money they earn and use. But it's a nice gesture to get people interested in their product.

Jeremy (11:15)
And you said it also a lot, I've heard you say it on this podcast and I've heard you say this with clients. There's a lot of free, really quality AI training out there. Yay Google, thank you for making courses available. ⁓ And look, if it helps you get the most out of Gemini, then fantastic. But there are lots of great.

Jim (11:32)
BT offers

them, they all offer them and then there are some that offer all across all things. We offer some of that too. But the point is definitely use the tools to learn more about the way to best use the tools. And that's the thing. What we stress is just do that because in the end, I will see you gonna learn this shit. You have to dive in and figure it out along the way.

Jeremy (11:52)
So ⁓ on today's tool tip, I wanted to dig in a little bit on how to use Gemini or really any LLM, right? Chat GPT or whatever, along with an office product in order to create something useful that you can share with your team.

Jim (12:06)
A small

win, but an early one and one that can be used regularly.

Jeremy (12:10)
Unlike some

of the other ones we've talked about this one actually is a little bit more It's not more technical but is more involved in that there's a lot of steps So yeah, so what I'll do is I'm just gonna talk about it now and then in the show notes I'll put a step by step. Maybe we'll do a follow-up So this is building a pod bot called a sales pulse board and I Well, it is now

Jim (12:26)
building a

I don't know if that's a term. I've

never used it in any of anything I've ever done. But okay. It's like a hot it's kind of like lead tracking and it's like sales track.

Jeremy (12:38)
Maybe it is.

It It's

sales tracking, but in the context of environmental conditions.

Jim (12:49)
things.

Yeah. Okay, so explain that a little bit. I definitely buy into this.

Jeremy (12:53)
Yeah, so you'll export your POS data, right? So point of sale. you, Jim. Your acronyms, yes. You'll export it as a... register system. Ready for another one? You'll export your POS data as a CSV, a comma separated values. is basically a spreadsheet, So export sales data for the week, and then import it into your LLM with a prompt.

Jim (13:11)
kind of spread.

be like Gemini or what?

Jeremy (13:19)
or

chat GPD or cloud or whatever. So you export your data set, you ⁓ import it, you write your prompt, then here's the bonus part, here's the environmental conditions. You look at things like what events are happening in your area. So if you're a restaurateur or if you're a retailer. ⁓

Jim (13:38)
or motels

or...

Jeremy (13:39)
Exactly.

So what are the other things happening in your hometown that might affect your sales or your revenue or whatever, but then also weather adding that into it as well.

Jim (13:48)
Weather being an external factor, events, I think of like fall, we're in the fall, so the idea of like there's an Oktoberfest in our town. It's like, hey, this Oktoberfest is going on, it goes on every year, it's a two-day event through a weekend every year in the same weekend, but the weather last year rained out both days, and so almost no one showed up. One day it was canceled. This year, beautiful weather, clearly you can't say it was the same, it wasn't. So Oktoberfest is a factor, a major event, external event, and then weather.

Jeremy (14:15)
Oh, your town is hosting a huge soccer tournament. Yay. Oh wait, but it's really raining. like what are the, and this kind of leads me to the last point about this. I think where it will really help you is if you save this data and then compare it, not month over month, but year over year. So same time last year versus this year, exactly. And then you'll get that data that will actually be useful. You can even modify your prompt in order to look at that year over year data, which actually this is my last point.

I think that any last point sorry the real value of this ⁓ is is obvious but I think the simplicity piece is going to be ⁓ enabled by turning whatever you make into a custom GPT right so that.

Jim (14:59)
So that means you can take it then. It hadn't handled it somewhat by itself, or at least in a slightly more automated way, less steps along the way.

Jeremy (15:01)
and then make it into a little mini app.

Even

just thinking about all the, we didn't even cover all the steps right now, but imagine that once all of that's done, you package it up and all you have to do in the future is export the data, attach it, click send, and then you get your report.

Jim (15:21)
or not eventually down the road if you're comfortable with it or you're comfortable with somebody doing it for you where you can add agentic AI where you can actually let these steps happen automatically to take those actions each week automatically without you or any of your staff having to think about it. Think about that for a minute where it's automatically happening and now fluidly just get this data and information and you can make good decisions based off of it and also not just predict for the future, look at the past but then predict for the future.

Jeremy (15:49)
Well,

so. I think actually it's it's quite easily achievable today if you're using athropic and it's connected to drive, you take your POS, you automatically week over week just have it dump your POS data. You're your point of sale data as a CSV into drive and then you just have the.

Jim (15:50)
for inventory, staffing and otherwise.

Everyone right? It's

so easy. We'll put it in the show notes. It's not as easy. He knows it's not as easy. It is easy. But if you kind of get in the swing of things and get in the rhythm, there are blocks.

Jeremy (16:18)
But more importantly,

don't you think, Jen, beyond how easy it is or not, whatever time investment you make in this, I think there's real value.

Jim (16:25)
Return

is there because now once you have it set and it's done right and it's tested Yeah Your automation is in place and you're golden and again all this is digitize all this stuff So you get away from those manual processes that you're doing in your head or by hand or on your phone and get them into? Systems that then can be automated

Jeremy (16:44)
And shared with team members.

Jim (16:46)
It's easier and easier

to do. We can all agree to that. It's going to get easier and easier to do. It is achievable now, but it will be even more achievable in the future, shortly in the next few months. And you might as well be ready, because if get ready now, you're to get even bigger dividends soon. Always.

Jeremy (16:59)
Watch out for the sea creep.