OpenAI Drops Plans To Release Adult Chatbot Amid Safety And Investor Concerns

OpenAI has shelved plans to launch an adult-focused chatbot, confirming that the project has been “indefinitely” put on hold, according to a report by The Financial Times.

The feature, initially unveiled in October 2025 with a planned release in December, had already faced delays as internal discussions continued over whether it should be launched at all.

Its cancellation marks the second product setback for the company this week, following the decision to discontinue its Sora video generator.

OpenAI Pauses ‘Citron Mode’ Release Over Safety And Research Concerns

The adult chatbot, reportedly known as “Citron mode”, currently has no timeline for release. Sources familiar with the matter told The Financial Times that OpenAI faced technical hurdles in adapting models that were originally designed to avoid erotic material.

The company also encountered difficulties in filtering out illegal content, including bestiality and incest, during development.

OpenAI said it intends to carry out longer-term research into the effects of erotic conversations and how users may form attachments to AI. It added that there is not yet sufficient “empirical evidence” to support a responsible rollout.

The firm further indicated it would prioritise its main productivity-focused offerings, such as coding tools, rather than pursuing experimental projects like the adult chatbot or Sora.

Investor Pressure And Internal Concerns Influence Decision

The proposal for adult features had followed earlier announcements that ChatGPT would include parental controls and automated age-detection systems.

Chief executive Sam Altman said in October that the company had historically approached such topics cautiously due to concerns about unhealthy reliance on AI, but believed it could “safely relax the restrictions in most cases.”

Despite this, the initiative reportedly raised concerns among investors, particularly after controversy surrounding rival xAI’s Grok chatbot, which was linked to the generation of deepfake explicit images involving real individuals, including minors.

There was also unease within OpenAI itself. One senior employee is said to have resigned over the issue, telling The Financial Times: “AI shouldn’t replace your friends or your family; you should have human connections.”

YOU MAY ALSO LIKE: Adobe CEO Shantanu Narayen To Step Down As AI Fears Shake Investor Confidence

Questions have also been raised about the reliability of OpenAI’s age-verification technology, which was introduced following legal complaints from families who claimed ChatGPT negatively affected their children.

Reports suggest the system may have an error rate exceeding ten per cent, potentially allowing underage users to access restricted features.

OpenAI said this level of accuracy falls within industry standards and noted that efforts are ongoing to improve performance.