π 5 min read
If you use ChatGPT for free, your browsing data is now being shared with advertisers – and that setting was turned on without your explicit consent.
On April 30, 2026, OpenAI sent an email to U.S. users announcing a major update to its privacy policy. The headline change: marketing cookies are now enabled by default for all free-tier ChatGPT accounts. Paid subscribers – Plus and Enterprise – are not affected by default. Everyone else is.
What Changed, Word for Word
WIRED compared the old and new policies side by side. The original text was unambiguous:
“We don’t ‘sell’ Personal Data or ‘share’ Personal Data for cross-contextual behavioral advertising, and we do not process Personal Data for ‘targeted advertising’ purposes.”
π§ Want more like this? Get our free The 2026 AI Playbook: 50 Ways AI is Making People Rich β Free for a limited time - going behind a paywall soon
That pledge is gone. It has been replaced with:
“Depending upon your choices, we may share limited data with select marketing partners for purposes of promoting our products and services to you on third-party properties. This is known as ‘targeted advertising’ or sharing for ‘cross-context behavioral advertising’ under certain state privacy laws.”
OpenAI also renamed the vendor disclosure category from “Vendors and Service Providers” to “Vendors, Service Providers, and Marketing Partners.” That small word addition carries significant legal meaning.
What Data Is Being Shared
OpenAI is explicit that conversation content is not shared. But everything around the conversation is fair game:
Join 2,400+ readers getting weekly AI insights
Free strategies, tool reviews, and money-making playbooks - straight to your inbox.
No spam. Unsubscribe anytime.
- Cookie IDs – tied to your browser and browsing behavior
- Device IDs – unique identifiers for your phone or computer
- Email addresses – sent to ad platforms to match you to ad impressions
The stated use case: if you see an OpenAI ad for Codex on Instagram and later sign up, the cookie or email match lets OpenAI confirm the conversion. Ad partners confirmed so far include Criteo and Smartly.
Free vs. Paid – The Two-Tier Privacy System
| Account Type | Marketing Tracking Default | Price |
|---|---|---|
| Free | ON by default | $0/month |
| Plus | OFF by default | $20/month |
| Enterprise | OFF by default | Custom pricing |
WIRED tested this directly – two free accounts had marketing settings turned on, two paid accounts did not. This is not an ambiguous opt-in. The default is set against free users.
The Money Behind the Move
This privacy shift did not happen in isolation. OpenAI has been building an ad business for months:
- January 2026: ChatGPT ads pilot launched in the U.S. – minimum advertiser spend of $200,000-$250,000
- February 2026: Ads began rolling out at the bottom of ChatGPT outputs for U.S. users
- Early April 2026: Ads Manager – described as similar in layout to Google Ads – quietly launched to pilot advertisers
- April 21, 2026: Cost-per-click ads launched at $3-$5 per click, alongside a CPM model where rates dropped from $60 to as low as $25
- May 1, 2026: EU compliance layer added to conversion tracking pixel, with country-level data handling and last-click attribution
OpenAI also hired David Dugan as its ads lead and is recruiting a Marketing Science Leader to own attribution models, incrementality testing, and media mix modeling. This is not a side project.
The context matters: OpenAI is reportedly burning through capital at a significant rate and is eyeing an IPO. A durable advertising revenue stream – one that doesn’t depend solely on subscriptions – makes the company substantially more attractive to public market investors.
How to Opt Out Right Now
If you have a free ChatGPT account and want to opt out, the path is buried but reachable:
- Log in to ChatGPT
- Click your profile icon, then Settings
- Navigate to Data Controls
- Find Marketing Privacy and toggle it off
OpenAI confirms users can opt out at any time. But the design choice to default free users into tracking – while defaulting paid users out – is a deliberate asymmetry. The company benefits from users who don’t notice or don’t bother.
What OpenAI Says
OpenAI spokesperson Taya Christianson told WIRED: “Nothing about our policy of not sharing people’s conversations or other private user content with advertisers has changed. Like many companies, OpenAI works with select marketing partners to help people learn about our products on third-party websites and apps, and we updated our privacy policy to clarify how this works.”
That framing – “clarify” – is generous. The prior policy explicitly banned targeted advertising. The new one explicitly permits it. That is not clarification. That is reversal.
The Broader Pattern
OpenAI is not unique in making privacy trade-offs for free users. Google, Meta, and virtually every major ad-supported platform operates on the same logic: the product is free because you are the product.
What makes this notable is the timing and the prior promise. OpenAI built significant user trust during a period when it explicitly distinguished itself from data-hungry ad platforms. That positioning is now being quietly retired as the company pivots toward monetization at scale.
For comparison, here is where the major AI platforms stand on advertising as of May 2026:
| Platform | Advertising | Free User Tracking |
|---|---|---|
| ChatGPT (OpenAI) | Active and expanding | Yes, default on |
| Gemini (Google) | Integrated with Google Ads ecosystem | Yes, Google account-level |
| Copilot (Microsoft) | No standalone ad program | Limited, Microsoft account |
| Claude (Anthropic) | None announced | No ad tracking |
| Grok (xAI) | X platform ads adjacent | X account-level |
BetOnAI Verdict
This is a significant but unsurprising move. OpenAI needs revenue beyond subscriptions to justify its valuation and fund the compute it requires to compete. Advertising is the logical path. Free users should not be surprised – ad-supported free tiers are the norm across the web.
What is legitimately worth criticizing: the opt-in default asymmetry between free and paid users, and the framing of this as a “clarification” when it is a policy reversal. OpenAI made an explicit promise. It is now breaking it.
What you should do: If you use the free tier and care about this, opt out using the steps above. If you use ChatGPT seriously for work, $20/month for Plus removes the tracking by default and gives you access to better models. That trade-off has always been the point.
The bigger question is whether this changes how users feel about the platform. Trust, once eroded, is expensive to rebuild. OpenAI is betting that 600 million weekly active users will not notice or not care. Based on how these things usually go, they are probably right.
Sources:
- WIRED – OpenAI Enables Marketing Cookies by Default for Free ChatGPT Users
- Digiday – OpenAI Starts Laying Foundations for ChatGPT Ads in EU
- OpenTools – OpenAI Flips the Switch: Free ChatGPT Users Now Tracked for Ads by Default
- PPC.land – OpenAI’s Privacy Policy Now Lets Advertisers Send Purchase Data
Enjoyed this? There's more where that came from.
Get the AI Playbook - 50 ways AI is making people money in 2026.
Free for a limited time.
Join 2,400+ subscribers. No spam ever.