Resources:
Join the discussion: Got something to say? Let us know on LinkedIn and network with other AI leaders
Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineup
Connect with Jordan Wilson: LinkedIn Profile
Why Personalized Ads in ChatGPT Are Inevitable: 9 Key Drivers Every Business Should Understand
Recent AI developments indicate a seismic shift in the digital advertising landscape. Personalized ads in ChatGPT are not only likely—they are essentially inevitable. Here are the exact reasons behind this transition, the underlying business value, and actionable insights stemming from the strategic initiatives described in the Everyday AI episode, "9 Reasons Why Personalized Ads Are Coming to ChatGPT Soon."
ChatGPT Personalized Ads: From Nonprofit Research to $500 Billion Market Cap
OpenAI's transformation from a nonprofit research lab into a $500 billion tech powerhouse is not superficial. The foundational pillars—ChatGPT’s memory features, its proprietary browser (Atlas), and a burgeoning commerce ecosystem—are intertwined to lay the groundwork for advanced advertising formats. While current digital advertising has been dominated by Meta (social) and Google (search), these legacy channels are losing ground as AI-powered conversational commerce becomes central.
OpenAI’s Burn Rate: The Critical Driver for Monetization
OpenAI’s annual cash burn is estimated to reach $8.5 billion, mainly due to the extraordinarily high compute costs required to deliver advanced AI services. The company’s user base surpasses 800 million weekly active users, with 95% on free plans. This user scale brings enormous opportunity—and necessity—for monetization. Sponsored content in the form of personalized ads is the direct path to sustaining the platform without throttling access or raising subscription prices.
ChatGPT User Data: Unlocking Unprecedented Marketing Precision
Unlike traditional keyword-based ads, ChatGPT's dialog-based experience enables deep data gathering. Every user conversation—be it a detailed business prompt or a personal question—generates a rich profile. Businesses can expect AI-powered ad targeting to reach granular levels of interest and intent by leveraging both explicit "memory" features and comprehensive conversation history. This leads to ads that match real-time needs, far exceeding the precision of traditional social or search advertising.
OpenAI’s Organizational DNA: Hiring Ad Industry Veterans
Nearly 20% of OpenAI staff now hail from Meta, including top-tier talent from Facebook’s ad team. These new hires bring proven experience in deploying large-scale, effective digital ad networks. The active search for a Chief of Monetization and the involvement of experts such as Fiji Simo (former Facebook ad lead) indicate that OpenAI is not experimenting—it is deliberately constructing a next-generation ad platform.
ChatGPT Memory and Browser Integration: The Technical Edge
The innovation spans beyond chat conversation data. Atlas, OpenAI's proprietary browser built on Chromium, creates a primary channel to capture comprehensive web usage patterns. When combined with ChatGPT Apps (integrations with services like Zillow or e-commerce), these systems synthesize user context across business, personal, and browsing activity. The data becomes even more valuable as sign-in with ChatGPT rolls out to other services, continually enriching the user profile.
Proactive Engagement: ChatGPT Pulse and Agentic Commerce Protocol
Recent introductions such as ChatGPT Pulse—a daily, context-aware news and data push—and the agentic commerce protocol with Stripe, which facilitates in-chat transactions and purchase recommendations, further set the stage for high-engagement, commerce-linked ads delivered directly within the user’s natural workflow.
These mechanisms support not only ad targeting but also instant, frictionless checkout experiences. This turns ChatGPT into a seamless commerce engine, with every promoted product or service tailored to the user's immediate situation.
Advertiser Value: From Top-of-Funnel Clicks to Deep Engagement
Traditional advertising spends heavily on acquiring leads. In contrast, conversational commerce as implemented in ChatGPT promises qualified engagement deeper in the funnel: users are interacting in context, often at or near decision points. Early adopter brands and agencies will see stronger click-through rates and higher conversions than on saturated platforms. Microsoft’s early conversational ad pilots in Copilot report a 73% improvement in click-through and a 16% higher conversion over search.
Privacy versus Utility: Strategic Considerations for Business
The depth of personalized advertising in ChatGPT is built on transparency and user trust. While some users may object to pervasive contextual ads, the business upside is significant. Businesses leveraging these capabilities must be prepared for a new wave of privacy and ethical discussions. The trade-off is clear: those opting for personalization yield far more relevant and timely offers, while also facing heightened scrutiny around responsible usage.
Action Steps: Seizing First-Mover Advantage
The transition to conversational commerce advertising is not theory—it is being rolled out in product, staff, and ecosystem strategy. First movers—whether enterprise or agile SMB—will find a less crowded competitive space, meaning lower cost-per-acquisition and superior access to qualified audiences. Investing in conversational ad strategies, preparing customer data for integration, and developing ethical tracking and transparency frameworks are recommended next steps.
Conclusion: Prepare Now for Personalized ChatGPT Advertising
Personalized advertising within ChatGPT is approaching mainstream reality. Its technical, organizational, and market momentum cannot be ignored. The winners will be the organizations adapting their marketing strategies to fit the conversational, context-aware, privacy-balanced ecosystem forming now. Early action is required to translate these emerging capabilities into bottom-line business outcomes.
Topics Covered in This Episode:
- ChatGPT Personalized Ads Hot Take Overview
- OpenAI's Burn Rate and Monetization Need
- Sam Altman's Shift on AI Advertising
- ChatGPT Free Users Data Monetization
- Memory Personalization Engine and Chat History
- OpenAI Staff Changes Boost Ad Strategy
- Atlas Browser User Data Control
- ChatGPT Apps and Cross-Platform Data Integration
- ChatGPT Pulse and Proactive Ad Delivery
- Agentic Commerce Protocol for In-Chat Purchases
- Affiliate Commissions and Ad Transparency Concerns
- Conversational Commerce vs Traditional Ads
- First Mover Advantage in ChatGPT Ad Platform
- Ethical Concerns with Personalized ChatGPT Ads
Episode Transcript
Jordan Wilson [00:00:16]:
It's not official yet, but it's my not so hot hot take that personalized ads based on your chat history are coming to ChatGPT very soon. You could almost say that's a given, but this is hot take Tuesday. So here's my hot take. I'm actually looking forward to personalized ChatGPT ads, and I think they'll ultimately be better for not only consumers, but also for businesses. And I'm obviously not making this proclamation empty handed, obviously. So on today's show, I'm gonna lay out the nine very convincing reasons personalized ads are coming to ChatGPT soon and also prepare you for what that means. Because I think it's gonna be a seismic shift that we all have to be ready for. Because the future of the Internet will be based on not traditional advertising, the ecosystem that Google and Meta have dominated for a decade plus.
Jordan Wilson [00:01:15]:
Now the future of the Internet is gonna be based on conversational commerce. And I think that OpenAI is in a prime position to be the leader. So let's get into the nine reasons why. Alright. Hope you're excited for today's show. I am. Let's get into it. Welcome to Everyday AI.
Jordan Wilson [00:01:31]:
My name is Jordan Wilson, and this is your daily livestream podcast and free daily newsletter helping everyday business leaders like you and me make sense of these nonstop AI updates, translate them into normal talk, and use that to grow our companies and our careers. That's what we do. It starts here with the unedited, unscripted livestream podcast. But if you wanna be the smartest person in AI at your company or your department, that happens on our website, youreverydayai.com. There, you can go sign up for our free daily newsletter. We're gonna be recapping not just the highlights from today's show, but also everything else happening in the world of AI because, yeah, today, there was a ton. I'm gonna say in the newsletter, I think it's probably been one of the busiest twenty four hours in AI news. Yeah.
Jordan Wilson [00:02:16]:
Amazon axing 30,000 jobs, a lot going on, so make sure to check out the newsletter. But by the end of today's show, stick with me for twenty five, thirty minutes. Let's see. But by the end of today's show, you'll understand why personalized chat GPT ads are inevitable. You're gonna know the nine tipping points that ensure it's actually happening, And you might even, by the end, agree with me that this is actually a good thing. It depends on what you think of personalized ads. Also, I put together a doozy for y'all. Alright.
Jordan Wilson [00:02:50]:
So, we put together a personalized ad playbook. This is one of those I should charge people for. I could, you know, sell this for a couple $100 on the website. But all you gotta do, if you're listening on the podcast, we always put in our show notes a link to the link in livestream. So go repost this, and we're gonna give you the personalized ad playbook, which includes a seven week conversational launch plan, high intent offers that don't feel like ads, and the executive readiness scorecard. Yeah. This one's a good one. Alright.
Jordan Wilson [00:03:21]:
So, let's get into it, though. What's happening? We have to zoom out first before we can zoom in to why Chattopty Ads are happening. Well, mainly, OpenAI is in the middle of a transition from a nonprofit research lab, which is what they were founded as, to a for profit company. Right? A five a company that's currently, has a market cap of $500,000,000,000, the last one. And I actually think it's probably closer to 600,000 based on some recent events. And it's kind of OpenAI's position is built on three things, and these, nine reasons are kind of based on these three pillars. It's memory, their new browser, and commerce. Those three things are very interconnected when it comes to the future of how the Internet's going to work.
Jordan Wilson [00:04:14]:
And this is a direct challenge to the only two players that I think are doing anything which is meta, handling digital advertising on the social side and Google handling it on the search engine and retargeting, remarketing. And you also have to understand, this goes along with kind of like almost this dead Internet theory, and the fact that SEO has been kind of dying or traditional SEO because so many people don't go to websites anymore. They just talk to chatbots about everything. Right? Which might be problematic. Right? Because you have people, you many millions of people using ChatGPT as an example, as a therapist. But that has created an interesting, future predicament for advertising. We also have to quickly look at CEO Sam Altman's shift on advertising because just over a year ago, he essentially detested ads. He was poo poo on ads.
Jordan Wilson [00:05:15]:
Right? He said that he hates ads, and he called them a last resort when asked if JAD GBT or OpenAI would ever use them. And he said the combination of AI plus ads is uniquely unsettling, but his tune has changed. So he did say that embracing ads could be potentially helpful and even a net win for users personally saying he saw some ads on Instagram that he found helpful. And why is this happening? Is it just because the CEO of, one of the largest companies in the world has had a, personal change of feelings towards ads? Well, you could make that argument, but no. That's not it. Well, what's different than a year ago or early twenty twenty four when these, comments first came is, well, now, OpenAI has raised tens of billions of dollars. And guess what happens when you raise tens of billions of dollars? As an example, SoftBank, one company, reportedly, $30,000,000,000, they've invested in OpenAI. You know what happens when you have many companies that have invested millions of many of billions of dollars into your company that is losing money? Well, you gotta have a plan, and you have to say, here's how we're going to start monetizing.
Jordan Wilson [00:06:36]:
And that plan, which I'm gonna get to later, involves a number. That number is 700 plus million. That is the number of free OpenAI or Chattopty subscribers that I think are gonna be seeing personalized ads soon. So we'll see if, paid users, those playing, paying 20 or, you you know, 25 for a team plan or $200 for a plus plan plan. We'll see if those users sorry. 200 for a pro plan. So we'll see if plus teams or pro users get ads, but it's inevitable. That's the hot take.
Jordan Wilson [00:07:17]:
It's coming, especially for free users. But I think it's actually a good thing. So it's not if, it's when, because they're happening. And here's the nine reasons I won't make you wait any longer. Number one, the burn rate. OpenAI, admittedly and reportedly, according to reports, is burning money. Up to, certain, publications have said 8 and a half billion dollars that they're burning through each year. Also, just kind of reference, Sam Altman's a 180 degree turn on ads.
Jordan Wilson [00:07:51]:
That's number two. Number three is, yeah, the number of users. They are straight up dominating the AI chatbot game with 800,000,000 weekly active users. Number four, the chat gbt memory personalization engine. Number five, the modification of OpenAI staff. We're gonna talk about the that information report that just came out. Number six, the Atlas browser having control of the data. Number seven, ChatGPT apps being able to, bring in new data and surfaces.
Jordan Wilson [00:08:23]:
Number eight, the agent, kind of checkout protocol, and the instant checkout. And then number nine, proactive delivery via CHGPT pulse. Alright. We're gonna go into each one of these a little bit more. But let's talk about recently. Those nine things, not all of them have been recent, but about half of them have just happened in the last, about five weeks. Yeah. So OpenAI, I don't think that this is, a secret anymore.
Jordan Wilson [00:08:57]:
Right? They haven't straight up said, yes. We are confirmed, bringing personalized ads. But they have said, whereas a year or two ago, they were a little mum on it, but they said that they're looking at it as an option in their hiring, head of monetization. So let's just look at the past five weeks. Right? OpenAI's full strategy has been laid out by their last couple of releases. So chat GBT Pulse, which is a daily proactive kind of push. It's only two people who are on the, the highest tiered plan right now, but it's essentially a highly personalized news delivery for you every day based on your context, your calendar, your information. Right? It is pushed to you.
Jordan Wilson [00:09:44]:
The second one, which came four days after, is OpenAI's agentic commerce protocol, which was, co developed with Stripe that allows agents to securely, check out online with a transaction record. Then in October, we had chat gbt apps a couple of weeks ago. This integrates new extremely valuable cross platform data, which for proper advertising, you need that. And then last week, we got maybe not the final piece to the puzzle, but definitely the biggest piece. Because if you seize control of the browser, you can own the entire experience across memory, browser, and checkout. So there you go. Those three things are all in place, especially with the announcement of OpenAI's agentic browser, Atlas. This was deliberate.
Jordan Wilson [00:10:37]:
This wasn't random haphazardly, thrown together go to market, products that were rolled out. We are gonna get official monetization announcements probably, I would say, by this year. Alright. Let's get into the nine reasons in a little more depth. Number one, the burn rate. OpenAI is burning cash. Alright? It's CEO a lot a lot of this is according to reports, but CEO Sam Altman has said that they're losing money on their hired priced plans. So OpenAI is projected to burn 8 and a half billion dollars this year alone just to keep the lights on.
Jordan Wilson [00:11:17]:
And that's driven by what they say is unbelievably expensive compute costs. Something goes with having some of the world's most powerful models. They're not cheap to run. Right? I am one of those people on the pro plan, and I guarantee you OpenAI is losing a ton of money on me, and they have been. Even though I pay for, like, five different accounts, six different accounts, something like that, they're still losing money on me because even as I talk, I got agents going right now. I got I got my my, Atlas browser open, and it's working. I have a scheduled task and scheduled agents going for me around the clock. Right? They're losing money.
Jordan Wilson [00:11:58]:
And the free user base yeah. And I've I've I've said this all the way back to 2023. Go back and look. I said even at the time, and this was, I think, like, GPT four. Right? This is wasn't even the good stuff. GPT four turbo, GPT four o. I said even that is worth thousands of dollars, right, a month. And it still is.
Jordan Wilson [00:12:22]:
The level of intelligence and the level of productivity that this unlocks for people that actually understand the technology and, teach their employees, it's hard to put a price tag on it. But the free user base has to be monetized. Right? Recent reports said that OpenAI's 70% of their total revenue, billions of dollars of revenue, right, comes from individual consumers, not from enterprise, not from the API, from individual consumers. So the thought is many people are gonna be mad, right, when personalized ads come to Chat GPT. It's like, alright. Well, if you want out of it, upgrade. Right? Whether the 20 tier won't have ads, I don't know. I'm not privy to that information.
Jordan Wilson [00:13:08]:
A lot of times I, you know, get get, you know, some information from my contacts at big, companies. This is not one of those. This is all just my own thoughts and beliefs from, you know, obviously talking about Checkatrade almost every single day for many years. But there's no other path to true monetization when you're already, number one, burning money. It's not a user problem. You got all the users, but this was intentional. Right? OpenAI, smart play, by the way. We mentioned this in our newsletter, I think last week.
Jordan Wilson [00:13:41]:
OpenAI and Anthropic two years ago, you know, year and a half ago, twenty months ago, we're very much could have gone down a similar path. I think Anthropic, prioritized being more profitable per user, and OpenAI prioritized users, which is the right move. Right? Because when you have 800,000,000 users, weekly active users, you can do what you want. You can turn a switch. And by that point, everyone's worked it into, into their daily processes. Alright. A little more on reason number one, the the just the money demands. They're huge.
Jordan Wilson [00:14:24]:
Right? We've seen a valuation. The last report said $500,000,000,000. So people I don't think I saw anyone else talk about this aside from me. Right? Right now, OpenAI would be a top 18 company in The US by market cap. Right? A lot of people think, oh, okay. You know, they're an AI startup. No. No.
Jordan Wilson [00:14:51]:
They're bigger, by market cap than almost every single company in The US and technically in the world. They are a tech giant. They are behemoth, but they are bleeding cash. Right? And investors like SoftBank eventually need to see some kind of return on that $30,000,000,000 of investment. And according to internal docs from OpenAI, they're not even expecting profitability on the current path until 2030, but they have to have this in place. Right? I'm sure somewhere there's a a GPT six pro model that's already done all this math and then saying, okay. Well, you know, we're gonna roll it out to this many users. This many users are gonna churn.
Jordan Wilson [00:15:33]:
This many users are gonna pay for it. Right? It's it it has to happen. The free users and maybe even the paid users are gonna be seeing ads, and companies are gonna pay a ridiculous amount, but we're gonna get to that later. Reason number two, internal forecasting. And this is obviously very much related to Sam Altman's one eighty, but there is no other way. Like I said, the single ad stream would be 20% of OpenAI's projected revenue if the free user monetization is forecast to scale. Reason number three. Ready for this number? 760,000,000.
Jordan Wilson [00:16:18]:
So, according to reports from this month, OpenAI, this is wild y'all. 95% of those 800,000,000 weekly active users as of this month, 95% of them are free users. Alright? So this is a potential pool of 760,000,000 people who are, number one, either gonna be getting ads that are highly targeted, insanely, personal, almost pervasive. Right? Yeah. Let's let's just call this spade a spade. Right? Think of all the things that people talk to chat g p t about. You don't put in those kind of keywords into Google. You don't put those things into meta.
Jordan Wilson [00:17:02]:
The conversations, the level of personal detail, the level of just spilling people's guts. Right? Whether it's from a business perspective, personal perspective, a big decision you're facing in life, even something as, you know, topical as planning a trip. Right? You're gonna be giving a lot more details to a large language model having an iterative conversation sharing details versus if you're just going to a search engine. Right? But that is 760,000,000 users who are likely either gonna be getting served ads. Talk about a huge pool for advertisers to tap in when current markets, specifically Google and Meta, those are your only two real big options, they are saturated, and they've been saturated for years. And this is how I think OpenAI plans to not just win ultimately against companies like Google and like Meta. This is their plan to win. But I think this is ultimately how they become profitable when they've raised that much money.
Jordan Wilson [00:18:04]:
They have bills to pay y'all, but being able to capitalize on that many users who are putting that much information, this is an unprecedented an unprecedented amount of data that OpenAI has to use to leverage and to monetize. Reason number four, the modification of OpenAI's DNA. So a recent report from the information which we shared about, both on our AI news recap show yesterday and in our newsletter last week. That's why you gotta be that's why you gotta be reading y'all. So the information reported that almost 20% of OpenAI's current staff, that's more than 600 employees, are former Meta employees. And a key hire who I think she is super, super sharp, by the way, Fiji Simo, the ex Facebook ad lead. Yeah. She's kind of responsible for, bringing ads to Facebook's news feed.
Jordan Wilson [00:19:04]:
Yeah. So now she's OpenAI CEO of applications. And I do expect that she's gonna be making some huge moves. And according to reports, she's actively hiring a chief of monetization now to build this ad platform. So, yeah, you see how the pieces are connected. Let's talk about reason number five. And we're gonna get kind of technical here, but I'm gonna try to zoom out. We all know memory.
Jordan Wilson [00:19:31]:
Well, if you use ChatGPT, you probably know memory. You see memory as this, this option or this feature inside of ChatGPT that brings along more personalization. As you're talking, ChatGPT might pull out certain things that you say. Right? I might say, hey. My name is Jordan. I'm from Chicago. I run a podcast called Everyday AI. We train companies on, you know, using ChatGPT Enterprise.
Jordan Wilson [00:19:54]:
Right? I might say all these things to ChatGPT, and it's gonna pull them out. These these little nuggets, and I can go in and I can delete them, turn them off, etcetera. That's one tier. That's saved memory. But then there's also chat history. Yeah. And that is where the real gold is because some people don't have memory on. Some people might delete things that they don't want ChatGPT, kind of remembering, maybe just for the case that they don't want it to show up in other chats.
Jordan Wilson [00:20:20]:
But chat history, that is a legit automated gold mine. This is profiling people probably better than an FBI profiler could because people are spilling their guts to chat GPT. We saw that when OpenAI, how they really, abruptly rolled out GPT five this summer, got rid of GPT four o, and the collective online world lost their noodles. Right? Because it was a very sycophantic model. Right? Jeep DeForest is like, oh, yes. You know? It's it's a yes, ma'am. Yes, sir. It just says yes to whatever you say.
Jordan Wilson [00:21:01]:
Right? And people, were using it in bad ways, but people became kind of addicted to it. Right? So they have this very long chat history where they're just sharing every single detail of their life, every problem they're facing, every scenario. Right? They're using it as a journal. Right? So OpenAI has this dual memory architecture of saved memories that are manageable by the user, but also a actual chat history. And that profile is huge. There's also that's where we get into some, some concerns both on the ethical side, the transparency side as well, because you technically don't know in that other layer. And I'm oversimplifying it by just saying chat history. It's much more technical than that.
Jordan Wilson [00:21:43]:
But there's some huge transparency concerns. And this, in theory, is every user's worst nightmare. Right? Because you probably don't, especially on a shared computer or if you're using Atlas. If Atlas just becomes your new family browser, you probably don't wanna see ads based on everything that you're chatting about. And here's where I'm going to differ from probably mostly everyone. Right? We put a poll out, in our newsletter yesterday asking people, hey. What do you think about ChatGPD personalized ads? And more than 80% of our audience essentially said some version of nope. Not good.
Jordan Wilson [00:22:21]:
I'm in the minority here. I'm in the 20%. I am absolutely looking forward to it. I'm gonna love it. I'm gonna probably click and buy every single thing that Chad GBT recommends for me. Right? Mainly, I'm using Chad GBT for my business. Not a lot on the personal life side, but I love highly targeted ads. Right? Yeah.
Jordan Wilson [00:22:39]:
I have a background in marketing and advertising. Sure. Did a lot of work with Nike and Jordan Brand back in the day, their marketing and advertising and comms teams. So I I I I love it from kind of a, you know, science meets art meets technology. I love that aspect of it. But even from a consumer standpoint, right, I'm not necessarily embarrassed about anything. I'm, you know, putting into chat GBT or googling. I I love being retargeted for things.
Jordan Wilson [00:23:06]:
So I know I'm in the, the minority there. But you have to think, especially if you're using ChatGPT for business. I think some people, you know, blur the line and they just use it as their life partner, life coach, life everything. Right? For me, I'm mainly just using it for business. Very rarely am I using it for anything personal. But even if I was, I wouldn't care. Right? I'm someone that's like, hey, big tech. Take all my data, but just make my life better.
Jordan Wilson [00:23:32]:
And I think this is one of those things. Right? If you look at traditional advertising, I don't think it's worked. I don't think it's worked. Right? Otherwise, advertisers would be much happier and consumers would be, I don't know, better off in their lives, but it it hasn't connected yet. But I think just the iterative conversation, that people have with ChatGPT. Right? And that's why ChatGPT also engages you. Right? So they have kind of these new features now where after you go back and forth with ChatChippity, ChatChippity is gonna ask you more questions by default. It didn't used to be like that.
Jordan Wilson [00:24:07]:
Right? It's kind of a feature, but, going after engagement. But what this enables, this dual memory architecture is it enables a level of targeting based on sensitive life events, you know, probably proprietary information maybe that you're sharing with chat g p d that you didn't know. Again, if you're on a free plan, because if if you're on a paid plan and have data sharing turned off, right, OpenAI hasn't been able been able to build this kind of profile on you. But if you're on a free plan, right, Let's call the spade a spade. If, you know, if you're on the free plan, that's that's that's that's your pay. That's your price of admission. It's OpenAI has been, and any company that you're using on a free plan, essentially, they've been building a profile on you just like it when you go into Google search. Right? They're collecting data.
Jordan Wilson [00:25:00]:
That's how the Internet works. That's how commerce works. That's how anything online works. But just the level of of sensitive information, life events, psychographics, everything. It goes far beyond keywords. What is capable for the level of personalization from ads based on conversations versus keywords? I I I can't even accurately describe, just the gigantic difference between the two. Alright. Reason number six, Atlas.
Jordan Wilson [00:25:32]:
You control the browser. You control everything. So it's very early. It's buggy. It's based on Chromium. So it's not like right? Google has completely dominated both the search market and the browser market for decades. But Atlas is based on Chromium, which is Chrome's open source project. So it should, in theory, be able to keep up, eventually with Chrome.
Jordan Wilson [00:26:03]:
Like I said, super buggy. I've been sharing, some bugs with the, the OpenAI team because I I love Atlas. But this seizes control of the user's entire web data stream and it bypasses Google. And its agent mode can autonomously complete multi step tasks like booking flights, like checking things. But, where where I think you also have to look at is there's a browser history as well. So now that dual kind of personalization or dual memory, now you have a third one. Right? So you have your memories in ChatGPT. You have your entire, chat history, but then you have your browser history as well if you are using the Atlas browser.
Jordan Wilson [00:26:44]:
Oh, and it gets even bigger because OpenAI released apps. And although there's only seven apps right now, when you log in and you authorize those apps, most people aren't reading the fine print. You're handing over that data. So if you go, connect Zillow, guess what OpenAI all of a sudden has? If you save things in your Zillow account and you're looking for, I don't know, a house in the Western Suburbs at $800,000, whatever, guess what OpenAI knows? Yeah. They they know a lot of things now, and they can start to triangulate that with everything else that you're talking about. And, again, the level of creepily maybe helpful, maybe not, but creepily accurate ads that they can spit out, mind boggling. Right? So bringing in this app, these ChatGPT apps, it shares the context of all of these different apps and everything that you've been doing inside of there. And I've been talking for a long time about the OAuth or, you know, sign in as ChatGPT.
Jordan Wilson [00:27:50]:
That's another thing that's gonna be rolling out to other websites as well. They've technically only done it within their own properties, right, on Sora. Right? But I think sign in with ChatGPT is gonna be rolling out to a ton of online services. Same thing. It's the same thing as apps, just on a different end. It's bringing in so much valuable app context and better ads that they can serve you. Alright? We have two more. Pulse.
Jordan Wilson [00:28:17]:
ChatGPT pulse. So a lot of people well, the reason why no one's talking about pulse is because it's still one of the few, features or modes of ChatGPT that is only available to pro subscribers. So those on the $200 a month plan. I'm not personally a huge fan of, Pulse yet. I think the product could be improved. It's not that good, at least for me personally. I'm an extreme power user of TidGPD. I think for your average right? Which okay.
Jordan Wilson [00:28:50]:
What people paying $200 a month are average users, but I think the average $200 a month user will find a ton of value. Me, not so much. But the concept is every single day, based on your data, based on what you're chatting about, based on your chat history, based on your memories, based on your your calendar, if you've connected it inside Chat GP like I have. It delivers you a proactive, it gets delivered in the morning, a proactive, essentially, newsletter. Right? If you've used like Apple News and you can, you know, flip through, I mean, the UI, the UX is great. I think for some people, it's gonna be very helpful. For me, not so much. But that is when that came out, I said this is the vehicle for personalized advertising.
Jordan Wilson [00:29:37]:
They're they're gonna be rolling that out to everyone. I think it's a a compute thing right now. They don't have the compute, to roll that out to, even the $20 plus and probably definitely not the free because I think that there's some heavy, agentic researching capabilities going on there. It also delivers, somewhat personalized images as well. So I'm guessing it takes a ton of compute, a ton of costs, but that is, I think, gonna be one of the more effective delivery mechanisms. Because once people are already used to Pulse, they're like, okay. Well, this is normal. It's normal for me to get something delivered to me like an ad.
Jordan Wilson [00:30:18]:
Right? That's highly personalized. And I think the lines are gonna start to blur between the two. More on that here in a minute. But what what, Pulse does is it does create just a habit loop, and it is prime real estate for future sponsored content. And then last but not least, we have the agentic commerce protocol, released just about a month ago, with Stripe as a co developer. And this what this does is it enables instant in chat purchases. So right now, there's not a ton of retailers, but there's some big ones, Etsy, Walmart, Shopify. It's actually hard to trigger one of these.
Jordan Wilson [00:30:54]:
So it's not like you can just filter. Right? Show me only things I can buy in app. Right? I've I've done a lot of prompts to get it to fire. So I've gotten it to fire. I haven't bought anything because I don't need to buy anything from Etsy or Walmart, right now. Me, I'm a I'm a Tarjay guy. But this is where OpenAI is all took ultimately gonna be making bank as well. Right? So this is gonna they're gonna be making money on free users, because this got rolled out to free and paid users.
Jordan Wilson [00:31:23]:
But OpenAI does earn affiliate commissions. So, there's some some some tricky things that they're not tricky in a in a bad way, but they're they're gonna be making a lot of money on that. Right? And we'll see, ultimately, I don't know how, this is gonna be measured or studied. But I think you're gonna start to see some people maybe complain, and I've noticed it a little bit myself. Right? When I'm using Chajipity for shopping and some of it, I'm using it for tests. And, you know, I'll be searching for something and it's I'm like, it's showing me a lot of Walmart results. And this is probably something I would not consider to be buying from Walmart. Right? A lot of tech gadgets.
Jordan Wilson [00:32:01]:
I'm like, I don't know. To me, this would seem more like a Best Buy type thing. Right? But I'm seeing Walmart, Walmart, Walmart. So I'm pretty sure there's gonna become, like, okay. Is this actually what the large language model would be showing me, or is this just a way that OpenAI is maybe pushing things that they're making money on? And will they eventually disclose that? Because it's not an ad. Right? It's just an affiliate, commission. So it's not like, you know, when you see ads eventually in your chatty bitty feed, there's gonna be an ad because you have to have that disclosure legally here in The US. But for an affiliate commission, I'm not sure.
Jordan Wilson [00:32:38]:
We'll see. And that brings me to one of the core conflicts. I think we thought of large language models as a helpful assistant. That is one of the first things in any OpenAI system prompt. Right? Is it helpful? It is a helpful assistant. But will that line start to blur? Now that people are maybe oversharing, and using ChatGPT for, I think, way more than OpenAI probably envisioned five years ago that people would be using it for. They're using it for every aspect of their life. And, you know, both with ads and with affiliate, commission responses that get pushed to the top, I think a lot of people are gonna be asking themselves eventually, okay, is this still a helpful assistant? Or is this a company now that has to make a lot of money because they have bills to pay? Right? And this isn't a knock against OpenAI.
Jordan Wilson [00:33:34]:
Like I said, I'm on the paid plan. I still want the personalized ads. I don't have a problem with the affiliate, commissions either. I do think those should be labeled. But at what point? I think one of the things that has helped, OpenAI's meteoric rise. Right? They've given people so much. So many features, very, very great plans, limits even on free base paid plans. It's you're getting a lot of technology.
Jordan Wilson [00:34:05]:
You're getting an insane amount of bang for the buck. I think OpenAI and and Google are the only two that you can say that about. So I think people have have this trust, and they built kind of a relationship or at least a rapport with ChatGPT and certain models more so than others. But at what point now will that start to blur the lines? Right? Is it just selling me something, or is it actually still advising me with my best interest in mind? But I think OpenAI needs to do, maybe a little bit better job of maintaining transparency, to preserve that long term trust. Like me, I'm researching projects, products. Why am I seeing so many things from Walmart now that I wouldn't think Walmart would be a top three retailer that I would look at for certain products? So even for me, I'm like, can I even trust these recommendations anymore? And here's the gist. Ads today, for the most part, are interruptive. Right? Even when you're searching something on traditional Google search, who does that anymore? Right? Those top one, two, three ads, everyone just scrolls scrolls by.
Jordan Wilson [00:35:19]:
They're in the way. They're intrusive. Conversational commerce, I don't think is gonna be, I don't think it's gonna be in our way. It's not gonna be intrusive. It's gonna feel natural. Right? It's going to feel like it's not an ad because it has, and and it will have so much context and so much, like such a deep insight into our personal and professional lives. I don't think they're gonna feel like ads. Where right now, even as good as Google advertising is or meta advertising is, it still feels like an ad.
Jordan Wilson [00:35:55]:
Right? It's it's in the way. You're scrolling social media. Oh, there's an ad. It's intrusive. Conversational? No. Because you are initiating it. Right? It's it's not like aside from polls. Right? Let's let's pull that aside.
Jordan Wilson [00:36:08]:
But it's not like you're gonna open chat GBT and, you know, go to chat and all of a sudden you get a pop up ad to go buy a pair of socks from Walmart. I mean, hopefully we don't see that. But the deep personalization also will create much higher engagement and conversion. The way traditional marketing and advertising works is you put so much money, time and effort to get people, to get new customers, new potential clients in, and they generally fall in this kind of top of funnel. Right? It depends. Right? There's there's a lot of, different ways I could go down this. But for the most part, especially b to b, highly valuable products and services, you know, you're paying to get clicks or you're paying to get leads, and that's all they are. Right? They're top of the funnel.
Jordan Wilson [00:36:52]:
You still have to nurture, entertain, engage them for a long time. When we're talking about conversational commerce, for the most part, these are probably people that are gonna be way further down. So especially early adopters, I think are gonna be printing money with this. All right. So you're gonna have also a first mover advantage and Microsoft CoPilot and some of their early, testing in their kind of conversational commerce ads, showed a 73% improvement on the click through rate in a 16% conversion lift over traditional search ads. And that's just right? This is very early on, very early on search before you had all of these layers of this, of this onion. So we're going from keywords to context. We're going from interruptions to longitudinal context, psychographics, and inferred intent.
Jordan Wilson [00:37:51]:
Sometimes I think large language models know what you're thinking, what you want, or what you need, even if you don't. And the reason why and I kind of see this all the time. Right? I'm chatting with different large language models all day every day throughout the day. Sometimes I forget things, or maybe I'm not seeing a connection between these 50 different conversations I had with Chad GPT today. Right? Guess who will be able to understand that? Well, OpenAI as an advertising platform, but also the advertisers who are spending money, they're gonna be able to triangulate and connect all these dots that you may not even know. Right? And I think that marketers, right? So whether you're a small business owner in the c suite, I think this is going to change the playbook for how advertising works. It is a foundational change, and I do think companies who are on the edge are gonna have a first mover's advantage because the competition elsewhere, it is tight. Right? Like I said, advertising on Google and Meta, it's not been easy for five, ten years.
Jordan Wilson [00:38:58]:
It's crowded. First movers are gonna have a huge advantage on Chat GPT because Chat GPT personalized ads are coming. So the trade off is privacy versus utility. Right? It's these personalized experiences that rely on deep user data. But you also get into ethical concerns because like I said, and I'm gonna wrap this up here quickly, not everyone's like me. But this is where, as a society, I think this is almost gonna be a a small little turning point, at least in the digital world. Right? We're gonna have to start looking at these AI chatbots, and, OpenAI is not gonna be alone. This is going to become the norm for monetizing.
Jordan Wilson [00:39:45]:
We're gonna have to say, is this okay? What, what permissions should be in place? What restrictions should be in place? Or should there be none? Because I think the first version is gonna be kind of the wild west, and OpenAI is gonna be a driver. And this is is so much more. It means so much more for our personal and professional lives. I think there's so many ethical hurdles that we must tackle here, especially with the way that large language models have been maybe misused, but in good ways. Right? So many people, are finding help and relief from, you know, using, AI chatbots as, you know, life coach, as a therapist, as a psychiatrist, which personally, I don't think is a good idea, but people are finding, relief in positive outcomes in their life. But what happens then when that gets monetized? Well, that's another conversation for another day. But the end the end game here, when we talk about from a business perspective, this turns it into a two horse game. I don't think Meta is gonna come out on top here.
Jordan Wilson [00:40:48]:
We'll see what they release on the AI side, on the large language model side, but they're spending money. And it seems like they're not making much progress. I could be wrong on that. But I do think that OpenAI's integrated architecture covering those three, those three pillars, right? Memory, browser, checkout. I think that is something that will actually be able to challenge Google's dominance. I do think eventually Google will follow suit, but it could be quarters or years later, until Google does something like that. They're a much bigger ship. It's going to be much more difficult, to be able to do something like that.
Jordan Wilson [00:41:24]:
OpenAI still, well, they still run like a startup. They don't have, shareholders. Yes, they have people who have invested in the company. But, I think right now this contest is no longer about search results. It's owning the conversation and where the conversation happens. And that is the nine reasons why personalized ads are coming to Chat GbT soon whether you like it or not. But if you want to be prepared, if your company needs a little more information, we put that together for you. I think this is one of the better, kind of guides that we put together.
Jordan Wilson [00:42:05]:
This is the personalized ad playbook, a three in one toolkit for any companies launching or scaling AI ads. Alright? If you wanna get ahead, this is how you do it. Like I said, we could be charging a lot of money for this. We're not. It's free. I like sharing good information that's vetted. So this is a seven week conversational launch plan, a high intent offer that doesn't feel like ads, how to write them and some examples, and then the executive readiness scorecard. So if you want those things, all three of them, just go repost this on LinkedIn.
Jordan Wilson [00:42:40]:
If you're listening on the podcast, appreciate your support. This is a longer episode. I said 25 to 30. It went to 40. Sorry. But just go look in the show notes. Go find this post on LinkedIn, repost it, and I will send that over. So thank you for tuning in.
Jordan Wilson [00:42:53]:
Please go to your everydayai.com. Sign up for the free daily newsletter. Thanks for tuning in. See you back tomorrow and every day for more everyday AI. Thanks, y'all.