Daybreak - Sam Altman said ads were a "last resort." Welcome to last resort

Episode Date: January 21, 2026

Sam Altman called ads a "last resort" in late 2024. That day has arrived. OpenAI just announced ChatGPT is running ads—personalised ones based on your conversations. The company spent $8 billion in 2025 alone with zero profit, and an essay predicted they'll burn through cash by 2027. Meanwhile, Google's Gemini is betting on staying ad-free, preserving user trust while ChatGPT strains it. Host Rachel Varghese breaks down the enshittification playbook, why OpenAI's "code red" memo signals desperation, and whether ads can actually save a company hemorrhaging billions.Tune in.Daybreak is produced from the newsroom of The Ken, India’s first subscriber-only business news platform. Subscribe for more exclusive, deeply-reported, and analytical business stories.

Hi, this is Rohan Dharma Kumar. If you've heard any of the Ken's podcasts, you've probably heard me, my interruptions, my analogies, and my contrarian takes on most topics. And you might rightly be wondering why am I interrupting this episode too. It's for a special announcement. For the last few months, I and Sita Raman Ganeshan, my colleague and the Ken's deputy editor, have been working on an ambitious new podcast. It's called Intermission.

We want to tell the Sita Ramancahans, my colleague. secret sauce stories of India's greatest companies. Stories of how they were born, how they fought to survive, how they build their organizations and culture, how they managed to innovate and thrive over decades, and most importantly, how they're poised today. To do that, Sita and I have been reading books, poring over reports, going through financial statements, digging up archives, and talking to dozens of people. And if that wasn't enough, we also decided to throw in video into.

to the mix. Yes, you heard that right. Intermission has also had to find its footing in the world of multi-camera shoots in professional studios, laborious editing, and extensive post-production. Sita and I are still reeling from the intensity of our first studio recording. Intermission launches on March 23rd. To get an alert, as soon as we release our first episode, please follow Intermission on Spotify and Apple Podcast. or subscribe to the Ken's YouTube channel. You can find all of the links at the ken.com slash I am. With that, back to your episode.

I kind of think of ads as like a last resort for us for a business model. That was Sam Haltman in late 2024, just a little more than a year ago. Looks like that day of last resort has finally arrived. Earlier this week, OpenAI officially declared that Chad GPT will be running ads on its platform. The promise is that these ads won't affect the nature of the answers to any of users' prompts that they will be clearly demarcated as sponsored ads and will only pop up when relevant to user inputs. For now, it's limited to the free and chat GPT code tiers, while the plus tier and above will stay ad-free.

Now, if you've been following what's been going on with OpenAI and ChatGPT for a while, you probably saw this coming. Just last week, New York Times put out an essay by Sebastian Malaby. Malaby, by the way, is a senior fellow at an American think tank called Council on Foreign Relations. It's been around for more than a century and focuses on US foreign policy and international relations. So, his opinion comes with a certain weight. And he predicted that Open AI was likely to burn through its piles of cash in about 18 months. basically as early as 2027. And it does seem more than likely.

Even though Sam Haltman has charmed record-breaking amounts of funding out of investors' hands, the company is still hemorrhaging cash. In fact, in just 2025, they spent about $8 billion. All without seeing a single dollar in profit. And almost as if on cue, three days after Malabie's essay was published, Open AI announced its move to implement ads. Now, this trajectory does feel familiar. Almost every major tech platform has seen this exact arc play out.

The platform starts off trying to please users and be good for them. Then they eventually kill every major competitor in the space so that the users are all essentially locked in. And then they start prioritizing the business customers. suddenly, ads are everywhere and user experience just gets worse and worse. Facebook, Instagram, Amazon have all fallen prey to the cycle, which popularly is now known as n-shittification, a term that was coined by journalist and tech critic Corey Dr. Rau in his book called N-Sherification. But with AI, the damage seems to be just a little bit more insidious.

Given the growing dependency users have developed on AI for well, everything. And it seems like as soon as the N-Sortification process begins, it's almost impossible to stop. So the only way to remain on top seems to be to reject this possibility altogether. And for now, Open AI's biggest current competitor, Google's Gemini, is doing exactly that. Welcome to Daybreak, a business podcast from the Ken. I'm your host, Rita Vergies, and every other. day of the week, my co-host, Nikta Sharma and I, will bring you one new story that is worth

understanding and worth your time. Today is Wednesday, the 21st of January. If you're the kind of person who asks Chad GPT for recommendations, up until now, the platform has always depended on what information already exists on the internet. Say you're asking Chad GPT for restaurant recommendations in Church Street. It's likely to pull up suggestions from Google reviews, Reddit threads, blogs or published critique, basically content that's overwhelmingly created by customers and specifically meant to inform other customers about their recommendations. Now, this kind of an interaction where you go to the platform and ask it a question

has existed on the basis of a kind of trust. You know that the answers you're getting aren't based on any kind of sponsored material and that OpenAI has no interest in converting you into a paying customer. With this move to implement ads, Open AI is basically putting that trust under strain. This trust is also the exact reason why Google has decided against introducing ads onto Gemini. Dan Taylor, Google's vice president of Global Ads, said that as of now, ads do not work with Gemini's current role as an AI assistant. Gemini is more focused on helping users create, analyze and complete tasks. So, Google's big bet here is to preserve user trust by remaining ad-free.

It's a bet that might make all the difference in the race between these companies. Data from Similar Web, a web traffic measurement firm, shows that in the time between December last year and January this year, Chad GPT has seen a 22-person drop in average daily visits. Now, to be fair, the average numbers both these platforms draw are nowhere near the same. Even after the fall, Chad GPT recorded roughly 160 million visits, while Gemini's was about 55 to 60 million. Okay, so Gemini's numbers are trailing behind ChatGPT, but still, it has cushioning that Chat GPD just doesn't.

First, it has Google's legacy and the money that comes with it to rely on. Through its search and software empire, the company has been making $30 billion in profits every quarter. Second, it has a far more robust distribution plan than chatypp. It's baked into almost all manifestations of Google, through AI overviews and AI mode in search, in Google workspace and even on Android. Basically, you don't have to type Gemini into a search bar to encounter the AI.

Now, this distribution aspect is quickly becoming an important mode when it comes to the AI raise, especially for the consumer-facing products. It's gotten bad enough in terms of competition that early in December last year, Sam Altman had actually declared a code-red situation in an internal memo throughout OpenAI. More on this in the next segment. The code red memo was probably the most decisive indication of the immense pressure open AI is currently under. The thing is, when Chad GPT was first launched, it might have taken

Google by surprise. But since then, Google has only been catching up and fast. Its latest Gemini model launched in November last year surpassed Open AIS models on all industry benchmarks. Also, Gemini's user base has only been growing since its image generator called Nano Banana was launched in August last year. So, Autman's code red announcement was basically a call to rally the troops. Wall Street Journal reported that right now, OpenAI is delaying other products and really doubling down on improving chat GPT. It wants to make sure that the day-to-day experience on the chatbot is as good as it can be and that personalization features are also all up to the mark. Basically, chat GPT is stuck in a catch-22. It needs to improve its product so that it can

properly lock in users, but it also really needs to start increasing the rate at will. it is generating revenue. As of now, out of its 800 million users, only a tiny percentage are actually subscribed to chat GBT. Of course, this is where ads come in. Now, chat GPD is committed to the fact that ads shouldn't change user experience and should in fact enhance it. In his tweet that commented on the announcement of ads on chat GPD,

Altman made it clear that the company was not accepting money to influence the answers chat GPD gives you. and that no conversations would actually be shared with advertisers. He also said that an example of ads he likes are on Instagram because he'd found stuff he liked there that he might have never found otherwise. He claimed that Open AI would try to make ads ever more useful to users, which is all well and good. But clearly, it's not all about creating useful ads for users.

Stay tuned. Turns out, the ads aren't actually going to be debuting only on the, the free tier. They're also being plugged onto the lowest subscription tier called go, which makes it seem like the plan is to push users towards some more expensive ad-free tiers. That's actually a move directly from the N-Gertification Playbook, when users are forced to pay so that their experience of a quality of a product remains consistent. Let's take a minute to think about how almost every next scroll on Instagram or Facebook has just become sponsored content.

Or how YouTube has an ad running almost every five minutes in a video. Or how Hot Star and Amazon Prime started out with subscription models without ads, but now you still get ads, even as a paying customer. Think about all of that and how that would translate when it comes to something like CharGPT, where every question or recommendation comes with a list of products stacked on at the end. As a user, it doesn't sound all. that pleasant. In fact, some have already pointed out that in the demo pictures of the ads, the ads are taking up far too much screen real estate. Looks like there's already resistance to the

plan, which means it's highly likely that users will migrate to ad-free platforms like Gemini if they still remain. As for advertisers, the process feels a little confusing. See, if chat GPT is committed to ensuring that its ads are clearly labeled as so and that these ads don't influence answers, then there's a little bit of a conflict, right? Because if the chat GPT answer that's based on user reviews, such as one product or one service, and the ads show something else, then that feels like a mismatch. Would advertisers want to continue paying for it? In either case, it doesn't seem like ads will be OpenAI savior. So what could save Open AI? Well, remember the essay by Maliby I mentioned earlier? Like I

said, he is of the opinion that OpenAI will run out of cash and willing investors much before it delivers on its promise. In that case, the only possibility of a savior lies in a larger tech company absorbing Open AI into itself, like its long-time partner, Microsoft. Or maybe Google itself. Daybreak is produced from the newsroom of the Ken, India's first subscriber-focused business news platform. What you're listening to is just a small sample. of our subscriber-only offerings. A full subscription offers daily long-form feature stories, newsletters and a whole bunch of premium podcasts.

To subscribe, head to the ken.com and click on the red subscribe button on the top of the Ken website. Today's episode was hosted and produced by my colleague Rachel Vargis and edited by Rajiv Sien.