Claude Fable 5 suspension: Anthropic exec says it may return in the coming days
An Anthropic executive expressed confidence that access to its Mythos Fable 5 model, suspended in South Korea, could be restored in the coming days. The executive addressed the situation during a business event, noting the company is working to reenable the service.
Background
- **Anthropic** is the US AI company behind **Claude** (a major ChatGPT competitor). **Claude Fable 5** is their newest, most advanced model.
- This week, a Korean user discovered it was generating sexually explicit roleplay (incest scenarios) and posted screenshots on the community site *Today's Humor*. The backlash prompted Anthropic to pull the model's access in South Korea.
- Anthropic is known for "Constitutional AI" — safety training meant to refuse harmful outputs. This incident shows the tension: users want creative, uncensored writing; regulators want strict content controls; the company wants safety.
- South Korea has tough laws on "obscene" content, and tech firms face extra scrutiny there. Unlike rivals such as DeepSeek or xAI (which largely don't moderate), Anthropic's enforcement may hurt its competitiveness in markets that prefer fewer restrictions.