Controversy around price and whether ChatGPT can compete with the likes of Google and Apple in the AI-arms race has got marketers talking in the wake of OpenAI’s announcement.
It didn’t come as a surprise when OpenAI announced ads were coming to ChatGPT this month… despite CEO Sam Altman once vowing that ads would be the “last resort” for the platform.
The sheer scale and cost of running large language models (LLMs), coupled with most users unwilling to pay for premium subscriptions, meant that bringing in ads was inevitable. It’s the solution to monetisation.
It does also bring feelings of déjà vu. After all, search started ad-free, before evolving to bring in sponsored listings.
The Microsoft-backed company has reportedly started trialling ad formats with dozens of brands in a pilot programme, with advertisers being asked to commit under $1m each.
Free and Go subscribers will start to see ads being tested in early February.
But the launch of ads in ChatGPT got affiliates and marketers talking for several other reasons: pricing, trust and competition.
The pricing controversy
OpenAI is pricing ads on a cost-per-view basis rather than the traditional cost-per-click model used in search and social media. It is reportedly charging $60 CPM (cost-per-millie), which puts ChatGPT ads closer to premium TV inventory than other digital platforms.
For context, Meta typically charges between $10-20 CPM – meaning OpenAI’s price point is three times higher.
This pricing controversy centres on two things, trust and tracking, Justin Deaville, Managing Director at Receptional, tells me. The big issue for marketers is the “lack of conversion tracking – advertisers only get basic metrics like impressions and clicks, nothing beyond that”.
For George Stolton, Head of Performance Media at ROAST, this was expected and seems to be the “trend for any emerging ad platform”.
“The bigger players have had 20 years to develop a sophisticated measurement solution,” he says, although the industry is seeing ad platforms develop tools faster in recent years.
“While referral traffic from ChatGPT can convert well, there will be a lot of trust placed in the platform without visibility on prompts. This limits the potential for direct ad optimisation, leaning more towards page optimisation for query types than specific prompt targeting.
“Brands will be eager to be early adopters, as every c-suite member is going to be pushing for presence here ahead of competitors, but justifying performance gains at a high price point may mean lots of experiments are short lived without visible return on that investment,” adds Stolton.
But Deaville argues the price point could be justified. Receptional research shows that conversion rates from AI-referred traffic can be three times higher than traditional paid ads. However, he does caveat that “without proper tracking, it’s impossible to know or optimise”.
The trust dilemma
Trust is going to be crucial from both a consumer and business perspective.
“Users are frustrated that even the $8 per month Go tier includes ads, especially after OpenAI previously implied they wouldn’t go down this route,” Deaville highlights.
However, the company has committed to keeping conversations within ChatGPT private and not selling users’ data to advertisers. This is partly why the insights are limited compared to Google and Meta, although OpenAI hasn’t ruled out offering more granular metrics in the future.
But the real issue for now is if the existence of ads then starts to interfere with user experience, such as if they start to feel like they are being nudged rather than helped. This introduces a “new kind of tension” for marketers, says Michael Norris, chief marketing officer at Youtech.
“Right now, visibility in AI answers is mostly earned through authority and usefulness, though the algorithm isn’t as tight as Google’s and can still be manipulated. This is an issue they’re already trying to tighten up.”
OpenAI is supposedly managing placements directly for now, however self-service campaign tools are in development.
“[But] if paid placements start influencing outputs in a meaningful way, brands won’t just be competing on relevance anymore, they’ll be competing on budget. Ideally, they’ll look to model Google’s approach and marry the two together,” adds Norris.
The big tech battlefield
OpenAI’s introduction to advertising comes at a time when other big tech players are upping their game in the AI-arms race.
Google has turned old search campaigns into campaigns that can surface ads in AI Mode using broad match and AI automation, explains Norris.
It will be interesting to see if OpenAI uses the same keyword targeting strategy or takes a different route entirely, especially as both platforms come in at the same place in the marketing funnel.
“From an SEO perspective, this doesn’t replace search, but it does further fragment discovery. We’re no longer talking about one primary surface. It’s search, AI answers, and potentially sponsored AI results all at once,” says Norris.
Add to that, Apple and Google recently announced a joint partnership, which will see the next generation of Apple foundation models based on Google’s Gemini models and cloud technology.
A joint statement from the two tech giants said the deal will “help power future Apple Intelligence features, including a more personalised Siri”. Apple Intelligence will continue to run on its devices and Private Cloud Compute.
“With Apple and Google getting increasingly close around AI on the iPhone, the pressure on OpenAI only increases,” says Norris. Therefore, against that backdrop, “experimenting with pricing and ads is expected” from OpenAI.
The pressure is certainly on. OpenAI has to delicately balance monetisation with user experience, ensuring ads don’t interfere with organic usefulness. Whereas legacy players like Google and Apple have already secured that tightrope.
Let’s not also forget that Google and Apple have a long-standing relationship. Google has paid Apple billions of dollars to be the default search engine on Safari for decades (dating back to the early 2000s). The Alphabet-owned company also just released an upgrade to make Gemini 3 the default model in AI Overviews.
There is an argument to say this new partnership and updates is Google’s play to continue being the default search engine… it certainly feels like a déjà vu moment.
But does that mean it’s game over for OpenAI even before its truly begun rolling out its advertising arm? I don’t think so. There is still room for competition, but what this suggests is perhaps collaboration among the big tech companies is the key.
The AI battlefield is heating up. But the race hasn’t quite finished.
For marketers, it’s all about authority. As Norris puts it, those that have real authority will benefit regardless of the channel in the age of AI search engines.