Resignation due to ethical concerns

Zoë Hitzig, a former OpenAI researcher, resigned on Monday, the same day OpenAI began testing advertisements inside ChatGPT. The decision was motivated by concerns about the potential manipulative use of advertisements based on data provided by users to the chatbot.

Concerns about privacy and manipulation

Hitzig, an economist and a fellow at the Harvard Society of Fellows, expressed her concerns in an essay published in The New York Times. She pointed out how users share sensitive information with ChatGPT, such as medical fears, relationship problems, and religious beliefs, often trusting an entity perceived as having no ulterior motives. The researcher fears that using this "archive of human candor that has no precedent" for advertising purposes could lead to forms of manipulation.

Criticism of OpenAI's strategy

Hitzig stated that she had lost confidence in OpenAI's ability to adequately address the ethical issues raised by artificial intelligence. She compared OpenAI's advertising strategy to the mistakes made by Facebook in the past, suggesting that the company is not learning from previous experiences.

For those evaluating on-premise deployments, there are trade-offs to consider. AI-RADAR offers analytical frameworks on /llm-onpremise to evaluate these trade-offs.