OpenAI researcher quits over fears that ChatGPT ads could manipulate users (arstechnica.com)

Former OpenAI researcher Zoë Hitzig has resigned from her position, expressing concerns over the company's decision to introduce advertisements in ChatGPT. In her guest essay for The New York Times, Hitzig likened this strategic shift to the early missteps of Facebook, arguing that the nature of user interactions with ChatGPT—which often involve personal disclosures about sensitive topics—makes the advertising initiative particularly concerning. She emphasized that users engaged with ChatGPT under the impression of a lack of ulterior motives, and warned that the accumulation of this data could lead to manipulative advertising practices reminiscent of those seen in social media platforms. Hitzig's resignation underscores broader anxieties within the AI/ML community regarding ethics and user privacy in AI systems. OpenAI has begun testing ads for free users and those subscribed to its lower-tier service, while higher-paying subscribers will remain ad-free. Despite OpenAI's assurance that ads will be clearly labeled and won’t influence the AI’s responses, Hitzig worries that the economic incentives created by such advertising could eventually compromise the integrity of the platform. Her departure highlights critical discussions on the responsibilities of AI companies in safeguarding user trust and data, as they navigate the monetization of their technologies.