OpenAI researcher quit over ads because she doesn't trust her former employer to keep its own promises

{ "@context": "https://schema.org", "@type": "NewsArticle", "headline": "OpenAI researcher quit over ads because she doesn't trust her former employer to keep its own promises", "datePublished": "2026-02-11T20:55:36+01:00", "dateModified": "2026-02-11T20:56:16+01:00", "author": { "@type": "Person", "name": "Matthias Bastian" }, "publisher": { "@type": "Organization", "name": "The Decoder" }, "description": "OpenAI wants to put ads in ChatGPT and former researcher Zoe Hitzig says that's a dangerous move. She spent two years at the company and doesn't believe OpenAI can resist the temptation to exploit its users' most personal conversations.", "articleBody": "Former OpenAI researcher Zoe Hitzig resigned over the company's move to bring advertising into ChatGPT. In a commentary for the New York Times, she makes one thing clear above all: she doesn't trust her former employer.\r\n\r\nHitzig doesn't consider advertising fundamentally wrong, but she warns that users have shared deeply personal things with ChatGPT: medical fears, relationship problems, religious beliefs. Turning that archive into an advertising tool creates a serious risk of manipulation.\r\nFor several years, ChatGPT users have generated an archive of human candor that has no precedent, in part because people believed they were talking to something that had no ulterior agenda.\r\n\r\nZo\u00eb Hitzig\r\nHitzig draws parallels to Facebook, which initially promised strong data protections and then gradually weakened those promises under pressure from its advertising model. Google's search ads tell a similar story - they've grown noticeably more intrusive quarter by quarter over the years. OpenAI is already optimizing for metrics like user engagement and making the chatbot more flattering despite internal warnings. Hitzig spent two years at OpenAI working on AI models and safety guidelines.\r\n\r\nOpenAI CEO Sam Altman has previously described Hitzig's scenario as a dystopia, so he's clearly aware of the risk. When the company launched its advertising test, OpenAI promised that ChatGPT ads would always be clearly separated from the chatbot's content. But Hitzig doesn't buy it.\r\nI believe the first iteration of ads will probably follow those principles. But I\u2019m worried subsequent iterations won\u2019t, because the company is building an economic engine that creates strong incentives to override its own rules.\r\n\r\nZo\u00eb Hitzig\r\nOpenAI is expected to go public later this year, which would ramp up pressure for fast revenue growth, especially given already inflated AI valuations. If nothing else, the debate should ensure that OpenAI's ad practices get plenty of scrutiny going forward.\r\n\r\nAs alternatives to ads, Hitzig suggests cross-subsidies from business customers, independent oversight bodies with real decision-making power over data usage, and data cooperatives modeled after the Swiss system.", "isAccessibleForFree": "False", "hasPart": { "@type": "WebPageElement", "isAccessibleForFree": "False", "cssSelector": ".paywall-active" } }