🔱 Sign up for a free demo today to see how the latest AI and digital advocacy tools can transform your work.
3 Minute Read
Think about the last time you asked ChatGPT, Claude, or Perplexity a question. Did you see an ad? Of course not. But that is almost certainly about to change—at least in some form. And when it does, it could fundamentally reshape how public affairs professionals reach audiences. In today's short read, we discuss why large language models (LLMs) will likely soon open the door to some form of advertising, and what this tectonic shift could mean for your work.
Until now, the LLMs have remained pristine—no banner ads, no sponsored results, no commercial interruptions. But the economics of that approach are becoming unsustainable. Training and running these AI models costs billions. OpenAI (the owner of ChatGPT) is expected to lose $8.5 BILLION in 2025. While subscription revenue helps, it's not enough to sustain the massive computational, energy, staff, and other overhead demands of next-generation AI.
Investor pressure regarding these massive losses is beginning to mount. In a sign that the LLMs are feeling this pressure, they are increasingly turning to off balance-sheet vehicles to build and maintain the massive infrastructure that they need to keep growing.
If you haven't heard much about this topic, don't worry about re-examining your media consumption. We see oddly little discussion of this issue in the financial and tech press, even by sophisticated commentators. But in our view, it's a simple equation. We live in the attention economy. The LLMs have our attention. Others want it—particularly advertisers. The LLMs need the revenue. Monetizing attention equals advertising.
.jpeg)
In fact, the LLMs have all but admitted that this is their plan. Several of the CEOs of these companies have openly mused on the topic. They have hired senior ad tech executives from other companies. And two of the major LLM owners (Google and Meta) are already giants in digital advertising.
Like Facebook before it allowed ads (their core business), the conventional wisdom is to not turn on ads until your user base has reached a critical mass. But that has clearly happened already in some cases. ChatGPT has over 200 million weekly active users. That's a larger audience than most traditional media platforms have ever achieved. In the crunch for revenue, the LLMs have an easy curtain to open that reveals dollar signs, and that curtain could open anytime.
When advertising comes to AI platforms, it's not likely to look like the ads you see today. Instead, it could take on several forms including:
Reselling Your Data To Advertisers: The LLMs could choose to keep their chat interfaces clean, but resell your data for advertising. This is the path that many data owners such as credit card companies, telecom companies, and financial institutions already take.
Sponsored Responses: When you ask a question about healthcare policy, an insurance company or policy think tank could sponsor a response that presents their perspective first. Similar to how Google provides sponsored responses to searches, the LLMs could offer sponsored responses to chats.
Contextual Messaging: Based on your chats, AI could identify when you are interested in a specific policy topic, and deliver a targeted "chat break," instead of today's "advertising breaks," from interested parties at precisely the right moment.
Before you say "why would I agree to any of this?," remember that the LLMs you use today are massively subsidizing your free or $20 per month subscription to build a critical mass of users. This subsidy is of course, the very definition of their operating losses. At some point, they will take that free lunch away, and offer you instead a "free," ad-supported version of your favorite chatbot, or an expensive monthly plan without advertising. Even if you choose the ad-free version, your audiences might not.
🔱 Get ready now to be a first mover: The constant theme of this newsletter is that the best ideas in ad tech come first to the commercial sector, and are slow, if ever to come to public affairs. AI advertising is likely to push this point to the extreme. Much of our sector is woefully behind on AI adoption, and can't even imagine the idea of AI advertising. The first movers in this space will have an enormous advantage. So . . .
🔱 Build your AI literacy now. Make sure your team understands how these platforms work. The organizations that succeed in AI advertising will be those that understand the technology the deepest.
🔱 If you're enjoying this content, please consider forwarding this email to a colleague or friend.
🔱 If you're not already a subscriber, please sign up here to stay up to date on the latest developments in political technology.
🔱 If you have suggestions for future newsletters or any other feedback, we'd love to hear from you.
_1.png)