ChatGPT Now Has Ads. Here's What Actually Changes for You.
OpenAI started showing ads inside ChatGPT on February 9, 2026. If you're on the free tier or the $8/month Go plan, you've probably already seen them: a single sponsored unit sitting below a ChatGPT response, matched to whatever you were talking about.
This isn't a rumor or a beta buried in settings. It's live for logged-in adult users in the United States right now.
What the Ads Actually Look Like
The format is restrained, at least for now. You'll see one ad unit below a response when OpenAI's system decides there's a relevant match. The unit can feature one or more items from an advertiser. Every ad is labeled "Sponsored" and visually separated from ChatGPT's answer.
Think of it like a Google Shopping result appended to the bottom of a conversation turn. If you're asking about project management, you might see a tool recommendation. If you're researching recipes, expect meal kit ads.
Who Sees Them, Who Doesn't
The split is straightforward:
Ads appear for: - Free tier users - Go tier ($8/month)
No ads: - Plus ($20/month) - Pro ($200/month) - Business, Enterprise, and Education plans
If you're already paying $20 or more per month, nothing changes. If you're on free or Go, your conversations now generate ad revenue for OpenAI.
How OpenAI Targets the Ads
This is the part worth paying attention to. OpenAI matches ads using three signals:
- Your current conversation topic (what you're chatting about right now)
- Your past chats (your conversation history)
- Your previous ad interactions (what you've clicked or dismissed before)
OpenAI says advertisers don't see your individual conversations, only aggregate performance data like view and click counts. Ads are excluded from sensitive topics including health, mental health, and politics. And they don't appear for accounts flagged as under 18.
Your Privacy Controls
You do have some knobs to turn. OpenAI lets you:
- View your ad interaction history
- Clear that history at any time
- Dismiss individual ads
- See why a specific ad appeared
- Turn off personalized ad targeting (though you'll still see ads, just less targeted ones)
What you can't do: remove ads entirely on the free or Go tiers. That's the upgrade path OpenAI is counting on.
The Money Behind It
This isn't a casual experiment. OpenAI is charging advertisers a reported $60 CPM (cost per thousand impressions), roughly three times what Meta charges. The minimum buy is $200,000. Launch partners include ad giants Omnicom, WPP, and Dentsu, with brands like Adobe, Ford, and Audible among the first advertisers.
At those rates, OpenAI is positioning ChatGPT ads as premium inventory, not remnant fill. They're betting that a recommendation inside an AI conversation carries more weight than a banner on a website.
Anthropic's Counterpunch
The timing wasn't subtle. The same week ChatGPT ads went live, Anthropic aired Super Bowl commercials with the tagline "Ads are coming to AI. But not to Claude."
The spots showed chatbot conversations derailed by absurd ads: a dating site called "Golden Encounters" popping up while a user asked for advice about talking to his mom. It worked. Claude jumped to #7 on the Apple App Store (up from #41), daily active users rose 11%, and site visits climbed 6.5%.
Sam Altman called the campaign "clearly dishonest." Anthropic kept running the ads.
Two weeks later at the India AI Impact Summit on February 19, Altman and Anthropic CEO Dario Amodei stood side by side for a group photo with PM Modi and Google CEO Sundar Pichai. When Modi lifted hands for a group gesture, Altman and Amodei raised fists instead of clasping hands. The clip went viral.
Your Decision Matrix
Here's what this means if you use ChatGPT regularly:
Stay on free/Go if: You use ChatGPT casually and ads don't break your workflow. The ads are non-intrusive for now, and you can disable personalized targeting.
Upgrade to Plus ($20/mo) if: You use ChatGPT for serious work and the ad interruptions bother you. You also get GPT-5.2 Thinking, higher usage limits, and early access to new features.
Consider Claude (free tier) if: You want an ad-free AI assistant without paying. Claude's free tier has no ads (Anthropic has publicly committed to keeping it that way). The tradeoff: Claude's free tier has lower usage limits than ChatGPT's free tier.
Consider local models if: You want zero tracking, zero ads, and full privacy. Tools like Ollama with Llama 3 or Mistral models run entirely on your machine. The tradeoff: no cloud-scale reasoning, more setup friction.
What This Signals
OpenAI is making a bet that most free users will tolerate ads rather than pay. That's probably correct. But it also means ChatGPT's free tier is no longer a neutral tool; it's an ad-supported platform where your conversation topics inform what gets promoted to you.
For practitioners who depend on ChatGPT in their daily work, the calculus has shifted. The free tier was always a trial with limits. Now it's a trial with ads and limits. That's a fundamentally different product.
The real question isn't whether ads in ChatGPT are good or bad. It's whether the ad-supported model will creep into the paid tiers over time, the way it did with streaming services. OpenAI says it won't. For now, that's a promise, not a guarantee.
