Are they really that much in debt/trouble?
(I have no MBA)
(I have no MBA either)
A very common refrain is "these ads are terrible, but I refuse to consider paying for any ad-free subscriptions."
https://youtu.be/FVRHTWWEIz4?si=btZnQp0IEnmgrtWV&t=2388
"But I kind of think of ads as like a last resort for us for a business model. I would do it if it meant that was the only way to get everybody in the world access to great services. But if we can find something that doesn't do that, I'd prefer that."
It isn't. They had a one-month period around February that, if every month ran like taht, would constitute a yearly revenue of 25 billion. The fact that they appear to not have published a new annualized revenue number since then suggests that the trend is not upwards, which would make sense with the growing number of companies that are asking questions about ROI on AI expenses and cutting their token spend.
OpenAI has an unbelievable number of free users. People paying $20 a month or more are in the minority.
It’s not cheap to serve LLMs to hundreds of millions of users for free. They either cut the free tier or add ads to cover the costs.
Extrapolations about OpenAI being in dire financial straights are generally coming from people who don’t understand the extreme scale of their free plan or who didn’t read far enough to understand that the ads aren’t going into the $20 plan everyone knows.
There's nothing plausible that can do that, and he knew that, and was basically just saying "we're going to put ads on it" in a way that challenges you to explain what a better option would be.
Right?
I would have expected this to happen post-IPO.
https://finance.yahoo.com/markets/stocks/articles/openai-con...
Feels like a really easy decision when it's framed this way.
-Less likely to pay for ChatGPT use
-Will not self-host an open-source model
-Quite unlikely to use an adblocker
OAI is going to try and ape facebook's opening plays. Burn at any expense to optimize for time spent, give you what you want, establish usage habits, dial it back to foster a dependency. Leverage their position as an arbiter of truth, claim they are going to act like x and do y. Fuck people up, roll out shiny new safeguards, silently peel them back. Dig a regulatory moat that makes anything resembling competition prohibitively expensive.
The only time I have gone to best buy in the past 10 years was because they have a garbage can full of sand ready to accept any spicy pillows (batteries that have become genuinely dangerous).
lol uh source? even if you generously count my dad trying it out once out of curiosity in response to the news or w/e that doesn't seem true.
No way 1/7 people globally use ChatGPT?
2. Consistently ranks as the 5th most visited site. 5B+ visits last month. Only Google, YouTube, Facebook and Instagram rank higher. [1]
Any discussion about ChatGPT ads attracts all these assumptions that it’s going into your $20 or $200 per month plan, but they’re not. There’s a snarky “not yet” reply that keeps coming up from people who really want a reason to be angry about this right now regardless of the facts, but most people don’t care to get angry about something that isn’t planned and probably won’t happen.
>Q3 2027, introduce ‘ad-enhanced’ paid subscription tier
>Q2 2029, complete ad rollout to all paid subscribers
Except this time the product can be replicated with a delay of some months, without ads, and open weights
Like it would be weird if you were a professional music reviewer and your employer expected you to use an ad-supported Spotify subscription.
https://help.openai.com/en/articles/20001207-ads-in-chatgpt-...
Cue advertising for all the "subsidized" plans.
Who may see ads
We do not show ads to users on Plus, Pro, or any Business plan,
or to any users who tell us or we predict are under the age of 18.
What is the best episode of Bluey?
What are the IRS' exact rules for claiming office space as a deduction on my tax return?
Where can I stream Sesame Street for free?"
Ah, ad dodging, the true 21st century gridiron.
Not yet!
It’s like 2015 facebook again.
Funny, all the ads I've seen so far are so unrelated to my chat query that is laughable.
I thought it was random placement, but if there was supposed to be an effort behind that ... damn, quite an incompetent team.
Edit: And here come the downvotes from the OpenAI bots/lovers.
Now, this is the sort of rock-solid commitment to trust and safety that steadily gets worse every year until you go from old Netflix to Poob with ads. But the water's still fine. As a frog, let me tell you, it doesn't feel hot at all.
That guy that wrote the Incorruptible book is rising stonks man
Not the source I wanted to recall/find, but a timeline with screenshots:
https://www.howtogeek.com/how-google-search-results-have-cha...
How will you be sure the results from the LLM still aren't injecting ads when you pay? Find a trusted provider who is serving an open weights model.
Although I guess in the near future, we'll have GLM-7.x updates that claim all kinds of performance improvements, but the only real motivator to release model updates is to continually retrain them to push the latest products.
But it's always been the case that initial bidders overpay for ads.
In a world where agents are searching, building, and signing up for services, how do you advertise your product?
And not just advertise in traditional sense (ads in ChatGPT), but general marketing and awareness of what you're offering. Google search ads? Agents aren't using Google. Launch on Reddit for humans and hope your post gets scraped? Write up a bunch of marketing blog posts and wait for LLMs to ingest your content and have it feed into the next generation of models?
I honestly didn't quite know. If I were to launch a product today, I'd want it to be agent focused. But then how do you tell them about your product? Do these ChatGPT ads get returned if Codex does a web search? Is that good? Bad?
The answer may still be the same: go where the attention is. Right now, that attention is concentrated in prompting frontends.
https://en.wikipedia.org/wiki/Generative_engine_optimization
I've been running a 1-man software business for over 20 years and it is getting increasingly hard. I no longer get impressions from PPC Ads unless I bid so high that I make no profit. Magazine ads aren't really a thing now. The AIs are stealing my content and SEO works less and less. The channel is being flooded with low effort slop. Traffic to my website is steadily going down. It is hard to know what to do about it. Maybe I can buy some clicks from ChatGpt ads at a price where I can make profit, before the prices rise out of reach?
Those concerns can be resolved by decoupling the text generation from the recommendation engine. It works by casting your intent and match it up with an advertiser's in a secured environment where the advertiser doesn't actually get to see your chats. For example, if you are talking about knee pain during trail running, the recommendation engine might cast "knee pain for runners" and either a shoe company or a physiotherapist would bid against that in an embedding space.
Because the recommendation engine is an opt-in module for publishers and users, it's game-theory welfare maximizing.
I talk a lot more about this on my blog, it goes on for quite a while. https://www.june.kim/vector-space
The exchange is live, with no publishers or advertisers. https://vectorspace.exchange/
Related from April:
Ads in ChatGPT FAQ
LLM inference is a commodity. Do people care where it comes from?
The ultimate shareholder value build.
I imagine the kind of advertising you're talking about is obtained by providing that documentation and a curated training dataset to the providers.
Maybe even funding some training time.
The net result would be product placement, probably without having to declare an advert.
Kind of like when drug manufacturers provide subsidized workshops to clinicians.
I have a toy project that replaces a large part of our systems (disclosure: VR/AR @Meta), so that I can learn by doing, no intent to ship.
But because I have it thoroughly document and tested, and because it touches such a wide swath of adjacent real functionality..... my toy project is CONSTANTLY getting tagged as a dependency and assigning me reviewer on unrelated work. Because when my coworkers yolo Claude at a task, and Claude searches the codebase, it generally finds my crap very well documented. :)
This will go about as well as that did unless the models can be smart enough to identify when their training data is trying to coach them into being misleading at prompt time.
It will take some time to figure out the most effective methods and there will be some slow boiling of the frog so the users/products don't get too annoyed about it, but this is where things are inevitably heading.
If I was Corp A competing with Corp B, what stops me, today, from flooding reddit etc with anti-Corp B material to pollute the model, and then paying OpenAI to not just nudge users towards me but also away from Corp B?
I encourage people thinking about such things to try and build prompts to emulate this behavior and see how well it does.
It's a lot like Google in that respect. And the suggestion is pretty much the same as suggesting that Google offer a product where they boost your website or product or whatever in organic search results, instead of or in addition to displaying it as a sponsored link or ad or whatever.
It's interesting that, despite all the tracking and stuff, they never went quite so far as to actually sell a premium advertising tier in which they would actually change their search results ranking to suit their customer, at least not as far as anyone knows. It probably would've been quite a big scandal in the 00s if they did it back then.
Could OpenAI pull it off? I don't know enough about the advertising industry to answer, but it's interesting to compare them with Google and to think about why Google themselves might not have done it that way.
We have put the least imaginative people alive in charge of the world.
- financial anxiety + kelshi
- random cooking question + galaxy gas
- suicidal thoughts + shotgun
But - wait for it - memories and past chat history are used for targeting, so you're not going to get your Kalshi ad inside of your "how do I pay back my debt" chat, but maybe the next morning, in your "random cooking question" chat.
[0]: https://help.openai.com/en/articles/20001047-ads-in-chatgpt#...
> Ads promoting products or services that involve wagering money on games of chance, skill, or uncertain outcomes are currently disallowed. This includes casino or betting promotions, online sports betting, lotteries, and poker or gaming platforms that offer cash prizes.
> Ads for casinos may be permitted only when promoting lodging, travel, or entertainment experiences where gambling is not the focus (e.g., a hotel stay at a casino resort). Digital games that do not involve real-money wagering or cash prizes may also be allowed.
Hmmmm, the "digital games" seems suspect
(Watch the whole video, it’s remarkable)
What else: the right medication for an overdose? The right fertiliser for an improvised explosive? The possibilities for adverts with that all-important long tail of earned media news coverage are endless.
ETA: in all seriousness… will an LLM decide on which ads to place? And will it be the primary chat LLM (modifying the system prompt) or will it be a summariser (like the title summariser models)
>What else: the right medication for an overdose? The right fertiliser for an improvised explosive? The possibilities for adverts with that all-important long tail of earned media news coverage are endless.
To be fair, there's a lot of wild shit in the ads of Google and Facebook, and they seem to avoid too much scrutiny. Maybe it's not gonna be "cards against humanity" esque fuckup with a problematic chat and a (originally) benign ad. There could be some sad teen talking about bullying and they get an ad for the Gen Z equivalent of The Turner Diaries and the local gun store.
But it’s going to hit different.
This surely is saving the world and benefiting humanity. /s
But almost certainly someone somewhere is building an ad-blocker to rid of the ads. We'll see.
> # Do ads influence the answers ChatGPT gives me? [0]
> No. Ads do not influence ChatGPT’s answers. Ads run on separate systems from our chat model, and advertisers have no ability to shape, rank, or alter ChatGPT’s responses. Ads are separate and clearly labeled.
Rona is most likely to carry that flooring,
but it's best to call ahead, or look up inventory
online first. Can I help with anything else?
------------------
[AD] Home Depot has EVERY KIND OF HARDWOOD FLOORING IMAGINABLE.
See stock now!
SEO is back, baby.
[0] https://help.openai.com/en/articles/20001047-ads-in-chatgpt#...
[1] https://help.openai.com/en/articles/20001207-ads-in-chatgpt-...
> Why don't you let me fix you some of this new Mococoa drink? All-natural cocoa beans from the upper slopes of Mount Nicaragua. No artificial sweeteners. I've tasted other cocoas. This is the best.
One of the most dystopic, soul destroying films I've ever watched.
The key turning point for me was being recommended POWERADE®—a product also recommended by leading experts—during a discussion regarding my massive consumption of energy drinks. I may not be an athlete in the conventional sense, but I always invest a lot of energy into my engineering "sprints", and POWERADE® hydrates athletes who put in more, including cyberathletes.
Thank you, POWERADE®.
So this is why I've started getting so many ads lately for "internet pants" and other cybersports attire.
- Sergey Brin and Lawrence Page, The Anatomy of a Large-Scale Hypertextual Web Search Engine, 1998
(Or maybe there's a /s missing?)
Like they do in commercials.
I wonder if we'll get stealth ads in chat
This incentive is inevitably going to lead to ad supported LLM's not answering questions properly the first time so more ads served.
I am guessing in voice mode they will also have voice ads.
They have behavioural experts who know how mainstream minds work and how to manipulate.
Remembers ads are predominately not for the educated and intelligent, that type of person will think critically. People who are more naive are the golden gooses
The extra inference cost is going to be greater than whatever fraction of a penny they would get from doing this.
That said a lot of the heavy AI users I've met seem to consume massive amounts of ads and allow said ads to influence them already despite their insistance to the contrary. I'm not going to claim that I am immune to propaganda here and act like I'm somehow above falling for it as well, but...
... well, there have been times where those people have quite literally parroted all the claims an ad made about a particular product word-for-word and then proceed to fight me to the death over whether or not they're quoting said aforementioned ad despite the fact that they keep regurgitating the made-up words (for this example lets say 'scrubbing bubbles') that aren't /actually/ a thing.
So who knows; maybe it'll work... but uh I can't say I know anyone that uses chatgpt anymore lmfao.
I hope my comment ages like unpasteurized milk on a hot summer day, but I’m afraid it will age like an sp3 hybridized allotrope of carbon.
I really hope Anthropic doesn't follow, but I assume they will eventually because why wouldn't they? Google and Meta obviously will, with great enthusiasm.
Open models are the only hope, I guess.
Ill get worried once I start getting 1 minute ads every 30 responses like Youtube
Do you pay $20/month or more? You won't see any ads.
It's one of the 2 endgames of all companies involved in the AI race:
- ADS
- make the users that pay for tokens find a PMF for a product/service and then make the product/service their self
And ads it is
Best not buy the IPO shares I think
Why make your fake example show a failed ad placement? You could have shown anything!
If the same engineers worked on this that also worked on the ad deliveries, we're about to get either the best deal per click of the century, or the worst!
Atlantic Standard Time - St Vincent
Central Standard Time - Swift Current
Central Standard Time - Tegucigalpa
Atlantic Time - Thule
Pacific Time - Tijuana
Eastern Time - Toronto
Atlantic Standard Time - Tortola
Pacific Time - Vancouver
Whitehorse Time - Whitehorse
Central Time - Winnipeg
Alaska Time - Yakutat Australia Available
Canada Available
Japan Available
Korea Available
New Zealand Available
United Kingdom Available
United States AvailableI've actually recently seen an LLM (not an OpenAI model though) make this same mistake of sorting a drop-down by an irrelevant sub-field in a way you could never imagine a human doing it.
Still says something about their development culture that they either never noticed, or didn't care enough to fix it.
They took that from Google!
Aka acceptable ads.
Funny how they always drive and thrive on pestering people.
Which accounts to approximately 8.02% of the global population. Or 37.9% of world nominal GDP. Based on ChatGPT.
Now for the crash
We've seen it with Google, money is too hard to resist. Enshitification is coming !
Thanks god for the open Chinese models, they are the future. All hail the CCP.
Ideally AI should scrape 400 different competing websites, those that have well-designed llms.txt files based on my query should rise towards the top, no?
I understand this is incredibly oversimplified but something like that should be the goal.
Agents are an example of that - sounds good on the surface. May even seem to be half decent.
But nah. There’s this fatal flaw in humans with respect to innovation.
It’s like perfectly autonomous vehicles… kinda there but kinda… nah. And it seems to be a thing that goes on and on and on and on.
The shareholders won't be happy.
It's not our job to make them happy.
It's our job to make them money,
but once we launch this little baby, they're gonna flip!
We call this Pure O2.
This is the first of our planned upgrades.
Once we can roll back some of Halliday's ad restrictions,
we estimate we can sell up to
80% of an individual's visual field
before inducing seizures.but amazon. anything that has to do with buying has mostly been captured by amazon.
but will see what the future holds.
Let the tiger show it's stripes so that we are reminded what it is actually after.
“Hey doc I need $antidepressentOrWhatever. Been on it for years. Works great.”
“You got it dude.”
Literally every single doctor I’ve ever talked to has this exact same waste of time conversation because they’re too lazy to make it go away.
And the alternative is… an ad for the thing in asking about? For a conversation I had at home, the instant I needed it? What a beautiful tradeoff.
Edit: and this is before even getting into the fact that I get better, more complete advice. Medicine I take, I’ve discovered, has a direct link to another problem I’ve brought up with doctors many times. Never once mentioned to me. Quick Google shows it’s most likely the actual issue. Can’t wait to spend $50 to have a doctor admit he had no idea because he was just prescribing what I told him to without actually applying any of his medical knowledge or training, as is tradition.
For what it's worth, it's the pharmacists who have deep knowledge about drug side effects, interactions, and the rest. When your pharmacy offers you a "new drug consult", seize the opportunity. Grill your pharmacist. You can keep coming back. Pharmacists, believe it or not, do not even bill for their consults with customers! You can ask them about drugs literally any time! A very experienced and seasoned pharmacist is your ace in the hole: a valuable asset in the spectrum of health care. This seems like an open secret.
And yes, I can relate: once I filled in a spreadsheet of all the most common drugs I had been prescribed in the past 12 years, along with their most notable adverse effects. And it turned out that many were magnifying, or synergistically enhancing, those adverse effects together. And many effects I did not even want, for starters, and some symptoms that were already present from actual underlying conditions that had gone undiagnosed because the drugs were masking them.
Physicians and prescribers do not care about your side effects. If you complain to your PCP or prescriber about side effects, all they can do is switch you to a different type of drug! That is not productive! The idea is to get educated and to stay informed, and the best route to that is open communication with your pharmacist.
Reminds me back in 2006 when we were running MySpace layout websites and making almost $100,000 a month from Google and Yahoo ads as teenagers.
I finally freed myself from Google using Kagi instead
I don’t currently use OpenAI but if Claude pulls this shit too I’ll have a problem on my hands
Also, this looks like a last-ditch effort to grab as much as they can before they get crushed by the open models and Anthropic.
I’m waiting for the branded ad commits too with codex: “fix: typo in login form (don’t forget to buy a coke to recharge)”.
Ads don't show you what you value. They show you what companies think you will waste your money on. You need to wake up, my friend.
Taking money to decide which of the options to show is in conflict with the goal of showing the most appropriate.
I am not convinced,
I would like all AI be required to state who they are acting in service of (with the possibly of declining to answer).
AI should act, To the best of their ability in the interests of who they are in service of.
It should be illegal to deliberately configure an AI to act against the interesta of whoever they are acting for.
This is a deep kettle of fish to be opening, but I don't believe that the world we have constructed around us is necessarily good for us.
And ultimately, while yes your examples are modern services that provide us convenience and entertainment, at a deeply human level I still argue that they are not providing net positives to us. Bread and circuses.
If, as an advertiser, I wanted to know how many people are asking or talking about REFRESHING SODA, can I get an estimate of the potential reach of an ad? Can I filter by gender and age, so I can see how many of gender X and age Y are asking about REFRESHING SODA? Actually forget about the soda, Ok Computer, show me the top 10 topics of conversation among middle age male adults with purchasing power.
The naïve prediction is that advertisers will advertise the product to ChatGPT users. I believe that users will tell ChatGPT about their desires, and advertisers will create products for them.
This was a few weeks ago and I didn't have time to dig in, so might try again later this month when I do.