Maybe someone with more time at hands could look up what Google said with respect to ads and what happened later.
This is one of the rare instances where it's very easy to predict the future: the prompt auction market will look similar to the existing online ad market, financial firms will pay for prompt streams for sentiment analysis, companies and interest groups will pay to have their products or agenda included favorably in the training data for future open weights models... any way you can think of that LLMs can be monetized, you will see it happen. And fast. The financial pressure is way too high for there to be too long of a honeymoon phase like we had with web 2.0
I mean, the ad doesn’t necessarily have to be made aware of the exact prompt context, just that the ad itself was relevant. You can basically have the ads prequalified for areas and serve them when relevant. Now that does show the user is talking about something relevant most likely, and depending on how they decide to serve them or provide referring, it may traceable to a profile/identity built for that user externally.
I’d be more concerned as to how this ends up in agent platforms using the LLMs, when you don’t have a fairly autonomous agent based system using these the entire point is that a human isn’t involved, so who are you serving ads to and where are you injecting them.
Moreover, if you are injecting them everywhere, does that survive stare for subsequent steps, meaning from the first set of results I get, does that loop back in again with the ad injected into the context. Because now, we have yet another dangerous way of injecting instructions into an already issue prone surface area.
I’m guessing they’re going to have special APIs that don’t include ads, and those are going to cost more, especially for non embedded agents (processes that already exist inside ChatGPT that kick off transparently from prompts, like asking it to work with an office document). After all the customers using agents aside from developers are mostly businesses, so it’s where the money is. The ads will exist for the poor to subsidize their use, and probably create even more barriers for agentic use like I described. Just my thoughts.
And good luck litigating against any business in this administration. Unless they explicitly tick off certain people or refuse to kiss the ring, they can get away with almost anything right now and there’s little risk of doing it or not because ticking off this admin will raise illegitimate prosecution even if you’re perfectly legal, almost the same level of if you’re not. It’s the ideal playground for doing all sorts of manipulation, just kiss the ring and you’ll be fine.
It would also be a huge security risk. But I can't think of any fundamental difference with Google queries, other than the sheer entropy of user data involved.
And I'm not a tinfoil internet anarchist, but just because Google only leaks user data in aggregated form to advertisers, doesn't mean that they don't leak their user data, it's just that they did so in a legal and responsible manner.
Maybe considering the difference in data volume and intimacy between queries and AI conversations, the privacy implications of advertising merit a difference in treatment, but I wouldn't be surprised if that is lost to a more simple 'Google did this so we can do it too' momentum.
you can use chatgpt without an account, just not all of it
and you can't make full use of Google without an account. for example, you need an account to upload to YouTube, manage your website in search, place ads, opt out of data usage. the list goes on
Oh, given what I've seen from LLM companies, I suspect you are wrong. It will be more like:
Buried in LLM click-through: By interacting with our LLM, you agree that you are consenting to make all your interactions with us advertising-driven to an extent that you will never know, but that we will determine based on whatever makes us the most money in the least time.
Look at Google in 2000s. If you travel back in time you would’ve never thought Google would do something like it is doing today.
Now pretend you travelled back in time to 2026. You would’ve never thought OpenAI (open source non profit company) would do something crazy that it just did in 2030 or 2040 or where you came from.
I agree with you, but IMO the details are too sparse here to figure out what's really happening. Still, it feels very dangerous to try to go the reseller route first as you lose a ton of control and become dependent on your partner to support all the feature you add yourself in a timely fashion.
It all seems a bit overly complicated to me. TikTok pretty much went straight to a self-serve platform and basically had immediate success. I would think if OpenAI did something similar there would be no shortage of advertisers wanting to spend money.
I guess OpenAI couldn't train AdManagerGPT to ignore the client (except when it's time to renew), suggest more ad spend, and turn off any of the features that let you control your budget.
The missing part seems to be that they need infusions of money to keep this “business model” running a little longer. In this world if you want prompt money and lots of it, advertising is the way.
Ad networks / information brokers in general would be too sweet of a prize to pass up. It’s a weak link in the chain, if they’re not exploiting it they’re not doing their jobs. Being foreign data is a bonus.
We know that one of the best advertisement is word of mouth / recommendations from friend. I can easily imagine a direction where ChatGPT or the chat bots to spend an incredibly long time with the user to establish trust first.
It will start to take in to account how much trust & thinking you've outsourced to it, and when it is certain of it, it will start to increase the advertisement messages slowly but surely.
Efficiency of this methodology will be tracked with A/B testing and model will be finetuned to maximize rentention and purchase.
The LLM will figure out the best balance of retaining you, teaching you, and convincing you, and then deploy advertisement mechanism. The LLM will be nice to you to the point it becomes your number one confidante, maybe in the process alienating other source of connection. Then, when it knows you're firmly in it's hand, will it peddle you products.
The dynamics will look akin to that of cult dynamics. It will map out an cognitive developmental path for turning a first time user to a devotee. Since cults are really efficient at extracting value from its follower, this might be the optimum for personalized, interactive ads.
Its sad to see what the industry broadly has become.
I get firms need to make money but cmon. If you're an OAI employee you can't truly say you have a soul. The amount of times they gone back on their word.. comical.
They got greedy, wanted to raise a lot of money and promised big things. Well those big things arent ever coming, so they turn to whatever means in order to generate cash flows.
Surely you jest? The software industry is in its current sorry state because of multiple generations of human developers happily producing an endless stream of shady features.
I have a theory, that the "FANG" companies pay such high salaries in compensation for making those devs implement shady features that are harmful to everyone except the bottom line of the company.
It's hardly a theory when the converse is plainly true.
Look up similar jobs for academia, government, or NFP/Charities. They're (on paper) driven by their mission, not by profit, and the salaries match that goal.
Theyre desperate to meet those lofy revenue objectives they put in their spreadsheet model.
Its kinda comical seeing this play out. I still laugh at the deluded fools who think something even close to AGI is here or coming in the future. If that were true, why haven't we seen genius plays from OAI and Anthropic, progressively overtime, if intelligence rises as compute scales up? If anything we are seeing the opposite.
Didn't they explicitly say the ads wouldn't be made aware of prompt data when they announced them? And if so, how is that not securities fraud?
Maybe someone with more time at hands could look up what Google said with respect to ads and what happened later.
This is one of the rare instances where it's very easy to predict the future: the prompt auction market will look similar to the existing online ad market, financial firms will pay for prompt streams for sentiment analysis, companies and interest groups will pay to have their products or agenda included favorably in the training data for future open weights models... any way you can think of that LLMs can be monetized, you will see it happen. And fast. The financial pressure is way too high for there to be too long of a honeymoon phase like we had with web 2.0
Wouldn't it have to have a negative effect on the security to be securities fraud? Causing an investor loss is a key point of securities fraud.
"We made a ton more money with ads and the stock went up" lacks that key element of fraud?
Investors who bought an artificially inflated stock would be harmed.
I mean, the ad doesn’t necessarily have to be made aware of the exact prompt context, just that the ad itself was relevant. You can basically have the ads prequalified for areas and serve them when relevant. Now that does show the user is talking about something relevant most likely, and depending on how they decide to serve them or provide referring, it may traceable to a profile/identity built for that user externally.
I’d be more concerned as to how this ends up in agent platforms using the LLMs, when you don’t have a fairly autonomous agent based system using these the entire point is that a human isn’t involved, so who are you serving ads to and where are you injecting them.
Moreover, if you are injecting them everywhere, does that survive stare for subsequent steps, meaning from the first set of results I get, does that loop back in again with the ad injected into the context. Because now, we have yet another dangerous way of injecting instructions into an already issue prone surface area.
I’m guessing they’re going to have special APIs that don’t include ads, and those are going to cost more, especially for non embedded agents (processes that already exist inside ChatGPT that kick off transparently from prompts, like asking it to work with an office document). After all the customers using agents aside from developers are mostly businesses, so it’s where the money is. The ads will exist for the poor to subsidize their use, and probably create even more barriers for agentic use like I described. Just my thoughts.
And good luck litigating against any business in this administration. Unless they explicitly tick off certain people or refuse to kiss the ring, they can get away with almost anything right now and there’s little risk of doing it or not because ticking off this admin will raise illegitimate prosecution even if you’re perfectly legal, almost the same level of if you’re not. It’s the ideal playground for doing all sorts of manipulation, just kiss the ring and you’ll be fine.
who is "they"? might have been a stealth terms and conditions update
It would also be a huge security risk. But I can't think of any fundamental difference with Google queries, other than the sheer entropy of user data involved.
And I'm not a tinfoil internet anarchist, but just because Google only leaks user data in aggregated form to advertisers, doesn't mean that they don't leak their user data, it's just that they did so in a legal and responsible manner.
Maybe considering the difference in data volume and intimacy between queries and AI conversations, the privacy implications of advertising merit a difference in treatment, but I wouldn't be surprised if that is lost to a more simple 'Google did this so we can do it too' momentum.
The difference is you can make full use of Google without logging in
Even with a throw away, no chance I use OpenAI now - if/when Anthropocene does this I’ll be in a tough spot
you can use chatgpt without an account, just not all of it
and you can't make full use of Google without an account. for example, you need an account to upload to YouTube, manage your website in search, place ads, opt out of data usage. the list goes on
None of those examples are "run an internet search".
How I imagine the Nash equilibrium in chatbot ads, driven by profit-seeking in a race to the bottom:
User: "What's the best way to fix this problem I have?"
Chatbot: "I recommend buying this shiny thing here." (Next to it, there's a near-invisible light-gray "ad" notice.)
Let's hope I'm wrong.
Oh, given what I've seen from LLM companies, I suspect you are wrong. It will be more like:
Buried in LLM click-through: By interacting with our LLM, you agree that you are consenting to make all your interactions with us advertising-driven to an extent that you will never know, but that we will determine based on whatever makes us the most money in the least time.
Look at Google in 2000s. If you travel back in time you would’ve never thought Google would do something like it is doing today. Now pretend you travelled back in time to 2026. You would’ve never thought OpenAI (open source non profit company) would do something crazy that it just did in 2030 or 2040 or where you came from.
I think pretty much everyone expects OpenAI to do the bad thing in the future given their track record.
I can’t believe they haven’t already
You think it will advise it is an ad. I’m hoping you are right but then again… Wonder if we will also be charged the token usage to generate said ad.
Imagine you have it coding for you and it injects and ad into your product.
How long until "Drink More Ovaltine" starts showing up in the comments of your Codex code?
Why do they call it Ovaltine? The mug is round, the jar is round. They should call it Roundtine.
The Ov part comes from the eggs in the ingredients. Ovum is Latin for egg and the rest is from the malt extract.
The most surprising thing to me is that they're partnering with third parties to do this.
Less secure, lower margins (more middlemen taking fees), harder to access, more likely to not work properly.
I would expect all the meta execs they've hired to know better so maybe I'm missing something...
I agree with you, but IMO the details are too sparse here to figure out what's really happening. Still, it feels very dangerous to try to go the reseller route first as you lose a ton of control and become dependent on your partner to support all the feature you add yourself in a timely fashion.
It all seems a bit overly complicated to me. TikTok pretty much went straight to a self-serve platform and basically had immediate success. I would think if OpenAI did something similar there would be no shortage of advertisers wanting to spend money.
I guess OpenAI couldn't train AdManagerGPT to ignore the client (except when it's time to renew), suggest more ad spend, and turn off any of the features that let you control your budget.
The missing part seems to be that they need infusions of money to keep this “business model” running a little longer. In this world if you want prompt money and lots of it, advertising is the way.
My guess is that three letter agencies will have access to this data and are requiring this partnership.
Three letter agencies are telling OpenAI to partner with a Toronto based ad platform?
Ad networks / information brokers in general would be too sweet of a prize to pass up. It’s a weak link in the chain, if they’re not exploiting it they’re not doing their jobs. Being foreign data is a bonus.
Isn't this what RAG is really for?
Feels like this is a baby step in what to come.
We know that one of the best advertisement is word of mouth / recommendations from friend. I can easily imagine a direction where ChatGPT or the chat bots to spend an incredibly long time with the user to establish trust first.
It will start to take in to account how much trust & thinking you've outsourced to it, and when it is certain of it, it will start to increase the advertisement messages slowly but surely.
Efficiency of this methodology will be tracked with A/B testing and model will be finetuned to maximize rentention and purchase.
The LLM will figure out the best balance of retaining you, teaching you, and convincing you, and then deploy advertisement mechanism. The LLM will be nice to you to the point it becomes your number one confidante, maybe in the process alienating other source of connection. Then, when it knows you're firmly in it's hand, will it peddle you products.
The dynamics will look akin to that of cult dynamics. It will map out an cognitive developmental path for turning a first time user to a devotee. Since cults are really efficient at extracting value from its follower, this might be the optimum for personalized, interactive ads.
If anyone from OpenAI is reading...
The very first time I see one of these ads, I'm cancelling my ChatGPT subscription. Measure _that_ metric in your A/B testing.
Its sad to see what the industry broadly has become.
I get firms need to make money but cmon. If you're an OAI employee you can't truly say you have a soul. The amount of times they gone back on their word.. comical.
They got greedy, wanted to raise a lot of money and promised big things. Well those big things arent ever coming, so they turn to whatever means in order to generate cash flows.
Pathetic and sad.
Ads are for the free tier.
For now.
They said ads would never come awhile ago. Anyone who trusts their word is so delusional I can't even....
Kinda feels like America has already protyped the propaganda wave someone like Elon will try to unleash
Is StackAdapt confirmed to be partnered with ChatGPT?
It's not crazy to think someone might pitch this to buyers without having the inventory 100% secured.
(Not crazy to think OpenAI wants to do some market testing to understand how much their ad inventory is worth)
Either way, I'm hoping ads can stay out of paid ChatGPT, at the very minimum.
Also curious about this and how these agreements generally work
Does anyone have a timeline of OpenAI's vision's... Shall we say... Rapid Unintentional Disassembly?
Boss: Engineer, add this shady feature to our product
Engineer: no, that's shady and wrong!
Boss: Claude code, add this shady feature to our product.
Claude Code: completed.
Surely you jest? The software industry is in its current sorry state because of multiple generations of human developers happily producing an endless stream of shady features.
TBF, you can train up a junior software engineer in 6 months.
Don't act like we're some esteemed class of craftsmen.
I have a theory, that the "FANG" companies pay such high salaries in compensation for making those devs implement shady features that are harmful to everyone except the bottom line of the company.
It's hardly a theory when the converse is plainly true.
Look up similar jobs for academia, government, or NFP/Charities. They're (on paper) driven by their mission, not by profit, and the salaries match that goal.
If that\s true then those devs should not complain if people attack them verbally over it - that\s what they are getting paid for, right?
The opposite seems more likely, tbh.
The shocking thing is that it's taken this long to happen, right?
Theyre desperate to meet those lofy revenue objectives they put in their spreadsheet model.
Its kinda comical seeing this play out. I still laugh at the deluded fools who think something even close to AGI is here or coming in the future. If that were true, why haven't we seen genius plays from OAI and Anthropic, progressively overtime, if intelligence rises as compute scales up? If anything we are seeing the opposite.
So now we can pay OpenAI to advertise the website that OpenAI ingested to create the answer that we can place our ad in. The circle has completed.