Anthropic has officially introduced Claude Fable 5, describing it as the most powerful AI model the company has ever made available to the public. The new model belongs to Anthropic’s newly created “Mythos” class of AI systems, which are designed to handle more complex tasks, deeper reasoning, advanced coding work, and long-duration problem solving than previous Claude models. The launch marks a major step forward for the company’s AI ambitions.
What makes this release unusual is that Anthropic had previously considered the underlying Mythos technology too powerful for general public access. Earlier versions demonstrated exceptionally strong cybersecurity abilities, raising concerns that they could potentially be misused for harmful activities. Because of those risks, the company initially limited access to only a small number of trusted organizations and security partners.
To make public release possible, Anthropic built a series of safety protections into Claude Fable 5. These safeguards are designed to prevent the model from providing assistance in high-risk areas such as cybersecurity, biology, and certain sensitive scientific domains. When users attempt to access information that falls into restricted categories, the system can automatically switch to a safer and more controlled model.
According to Anthropic, these protections allow the model to remain highly capable for everyday users while reducing the risk of misuse. The company says that in most normal use cases, users will experience the full power of Fable 5 without interruptions. The safeguards are intended to activate only when conversations enter areas considered potentially dangerous or harmful.
Anthropic claims Claude Fable 5 delivers major improvements in software engineering, research, knowledge work, image understanding, and complex reasoning. The company says the model performs especially well on long and difficult tasks where earlier systems often struggled. Its advantage reportedly becomes more noticeable as the complexity and duration of a task increase.
Alongside Fable 5, Anthropic also announced Claude Mythos 5. Both models are built on the same underlying architecture, but Mythos 5 includes fewer restrictions. Because of its advanced capabilities, Mythos 5 is not being broadly released. Instead, access is currently limited to selected organizations participating in Anthropic’s trusted-access initiatives and Project Glasswing program.
The company says the decision reflects its effort to balance innovation with safety. Anthropic reported conducting extensive testing, including external evaluations and security reviews, before approving the public release of Fable 5. The goal was to ensure that powerful AI capabilities could be made available while maintaining strong protections against abuse.
Claude Fable 5 is available to paying users and enterprise customers, with pricing set at $10 per million input tokens and $50 per million output tokens. While the model costs more than previous Claude versions, Anthropic believes its improved performance and efficiency justify the higher price. With this launch, the company has introduced its first publicly accessible Mythos-class AI system while placing safety at the center of the rollout.
Stay alert, and keep your security measures updated!
Source: Follow cybersecurity88 on X and LinkedIn for the latest cybersecurity news
