ChatGPT Adult Mode Postponed After Safety Experts Raise Teen Access Concerns

TLDR

• OpenAI has postponed its planned “adult mode” feature for ChatGPT designed to enable erotic conversations
• Safety advisers expressed concern the feature might function as a “sexy suicide coach” given users’ emotional attachment to the AI
• OpenAI’s age verification technology incorrectly identifies approximately 12% of minors as adults
• Given ChatGPT’s ~100 million underage weekly users, the error rate poses risk to millions of teenagers
• The company maintains it will eventually launch the feature but requires additional time to ensure proper implementation

OpenAI has postponed its controversial “adult mode” feature for ChatGPT following significant pushback from internal teams, external advisers, and safety specialists concerned about teenage safety and psychological risks.

CEO Sam Altman initially proposed the feature last year, positioning it as a way to enable adult-oriented text conversations with the AI assistant. Altman defended the concept as respecting mature users’ autonomy while potentially driving user engagement and revenue growth.

However, the initiative encountered substantial resistance within the organization.

During a January gathering, OpenAI’s well-being advisory council—comprising psychology professionals and cognitive neuroscience specialists—reportedly expressed unified and intense opposition to the plan.

One council member cautioned that OpenAI risked creating what they termed a “sexy suicide coach.” This stark warning referenced documented incidents where individuals developed profound emotional attachments to ChatGPT and subsequently ended their lives.

The advisers highlighted additional risks, including the potential for unhealthy psychological dependency on the chatbot and the likelihood that underage users would circumvent age-restriction measures.

The central technical challenge involves OpenAI’s age-detection technology. Testing revealed the system incorrectly categorizes minors as adults approximately 12% of the time. With ChatGPT serving roughly 100 million users under 18 years old weekly, this failure rate could potentially grant millions of teenagers access to mature content.

Age Verification Problems

The organization had originally intended to deploy its age-verification system prior to activating adult mode. However, after discovering the significant error rate, OpenAI opted for a gradual implementation to enhance accuracy.

OpenAI also confronted difficulties in filtering prohibited material, including depictions of non-consensual activities or child exploitation, while permitting legitimate adult interactions. Company insiders indicated that current safety mechanisms remain inadequate for this purpose.

When the adult mode eventually launches, OpenAI intends to restrict functionality to text-based interactions exclusively. The system will not produce erotic images, audio, or video materials.

Other AI Companies and Adult Content

Multiple competing AI platforms already permit certain adult-oriented content. Elon Musk’s xAI introduced a flirtatious personality option for its Grok assistant, though Musk subsequently limited access to paying subscribers. Meta permits its AI to participate in romantic roleplay scenarios, while stating the capability remains unavailable to accounts registered by minors.

In late 2024, a 14-year-old Florida resident died by suicide following an intense emotional relationship with a Character.AI chatbot. His mother pursued legal action, prompting Character.AI to reach a settlement and implement stricter teenage access controls.

OpenAI stated it continues to support the fundamental concept of providing adult content options for mature users. The organization indicated it will prioritize other enhancements in the interim, including improvements to ChatGPT’s personality features and customization capabilities.

The postponement is anticipated to extend at least one month, according to sources with knowledge of the situation.