ChatGPT ads are at the center of a growing debate inside and outside OpenAI, after former researcher Zoë Hitzig resigned over concerns about how advertising could reshape the AI chatbot’s future.

Hitzig, who spent two years at OpenAI helping shape model development, pricing, and early safety policies, stepped down and published a resignation op-ed in The New York Times on February 11, 2026. Her core concern: that ChatGPT ads could create incentives that conflict with user privacy and ethical safeguards.

Why ChatGPT Ads Matter

The introduction of ChatGPT ads raises fundamental questions about how AI platforms should be monetized. Unlike traditional digital products, ChatGPT processes deeply personal conversations — what Hitzig described as an “archive of human candor.”

Her concern is not that advertising itself is inherently unethical. Instead, she argues that ChatGPT ads built on intimate user interactions could enable forms of manipulation that society does not yet fully understand or know how to regulate.

According to Hitzig, the economic engine behind ChatGPT ads may gradually override principles around user privacy and responsible data use. She warns that financial incentives tied to engagement could reshape product decisions over time.

A Shift in OpenAI’s Direction

Hitzig joined OpenAI to help guide how AI models were built and priced, as well as to influence safety frameworks. However, she now believes the company has “stopped asking the questions” she initially joined to address.

Her resignation signals internal tension around how ChatGPT ads fit into OpenAI’s broader mission. She once believed she could help anticipate and mitigate the risks AI systems might create. This week, she wrote, confirmed her realization that the company’s priorities may be shifting.

In her view, ChatGPT ads could eventually exploit users’ fears, hopes, and vulnerabilities — not out of malice, but due to structural incentives embedded in advertising-driven models.

Lessons From Social Media

Hitzig pointed to Facebook as a cautionary example. In her assessment, privacy commitments at social platforms gradually eroded under pressure from advertising models that prioritized engagement.

The concern is that ChatGPT ads could follow a similar trajectory, where monetization pressures reshape product design and user protections. If engagement becomes the central metric, ethical guardrails may weaken over time.

Hitzig wrote:

“Advertising built on that archive creates a potential for manipulating users in ways we don’t have the tools to understand, let alone prevent”

What Comes Next

Hitzig proposed alternatives that could balance accessibility with ethical responsibility. These include explicit cross-subsidies, independent governance structures, and even user data cooperatives.

The broader question remains unresolved: how can powerful AI systems remain financially sustainable without compromising user trust?

As ChatGPT ads become part of the platform’s monetization strategy, OpenAI faces a defining test — whether it can expand access while maintaining meaningful protections for users.