Anthropic launches Claude Fable 5: How it could change the way people work

Anthropic has launched Claude Fable 5, the first publicly available version of technology derived from its highly discussed Mythos AI system. The move marks a significant moment for the artificial intelligence industry because it gives developers, businesses and researchers access to some of Anthropic's most advanced capabilities while keeping strict safety protections in place. The company says Fable 5 is designed to perform complex software engineering, analytical work, reasoning tasks and vision-based operations. However, unlike the original Mythos model, which was restricted to a small group of partners, Fable 5 includes built-in limits for high-risk subjects such as cybersecurity, biology and chemistry. When users enter those areas, the system automatically switches to the safer Claude Opus 4.8 model.

What exactly is Claude Fable 5?

Claude Fable 5 is Anthropic's attempt to bring advanced AI capabilities to a wider audience without exposing the most sensitive parts of its technology. Mythos first attracted attention earlier this year after Anthropic claimed the model could identify thousands of software vulnerabilities and uncover security flaws at a speed far beyond traditional methods. Because of those capabilities, access was initially restricted to selected organisations managing critical infrastructure and cybersecurity systems. Now, Anthropic has created Fable 5 as a public-facing version of that technology. According to the company, most users will experience the full power of Fable 5 in more than 95 per cent of interactions, while only a small number of requests will be redirected to another model for safety reasons.

Why is Anthropic being cautious?

The release of Fable 5 comes at a time when AI companies are facing growing questions about safety, security and control. Anthropic has repeatedly warned that future AI systems could become significantly more capable and may eventually reach a stage where they can improve parts of their own performance with limited human guidance. Because of these concerns, the company says it carried out extensive testing before launching Fable 5. According to Anthropic, internal and external security researchers spent more than 1,000 hours attempting to bypass the model's protections. The company said testers failed to discover a universal method that could consistently defeat the safety systems. Anthropic has also introduced a 30-day data retention requirement for users of Fable 5 and Mythos 5. The company says the data will not be used to train future models but will help identify new threats, jailbreak attempts and unexpected behaviours.

How powerful is Claude Fable 5?

Early testing suggests Fable 5 performs strongly across a range of tasks. Analytics company Hex said Fable became the first AI model to achieve a score above 90 per cent on its benchmark for complex analytical work. Meanwhile, software platform Base44 reported improvements in application development, saying the model was particularly effective at building complete applications from a single prompt and handling tool integrations.

Another AI platform, Genspark, said Fable outperformed competing models in several tests involving user interface design, coding and agent-based tasks. According to the company, the model showed stronger judgement and attention to detail when handling complicated requests. These results suggest Anthropic is focusing not only on raw intelligence but also on practical business applications.

How could Fable 5 change the workplace?

The biggest impact may be felt in software development, business operations and knowledge work. For software engineers, Fable 5 could help write, review and debug code faster. Companies may use it to automate parts of application development, identify bugs and improve productivity. In some cases, businesses could complete projects with smaller teams and shorter timelines. Knowledge workers may also benefit from faster research, data analysis, report generation and decision support. Instead of spending hours reviewing documents or analysing information manually, employees could rely on AI to complete much of the groundwork. The model's ability to reflect on its own work and verify outputs may also improve reliability in areas where accuracy matters.

The cost question remains
Despite the excitement around the launch, Fable 5 will not be cheap.
Anthropic has priced both Fable 5 and Mythos 5 at $10 per million input tokens and $50 per million output tokens. That is roughly double the cost of Claude Opus 4.8. This comes at a time when many companies are becoming more careful about AI spending. Several businesses have reported using up annual AI budgets faster than expected as adoption increases across departments.

As a result, organisations may need to decide whether the performance gains offered by Fable 5 justify the additional expense.

What does this mean for the AI race?

The launch of Fable 5 highlights how quickly the competition between AI companies is intensifying. Anthropic, OpenAI, Google, Meta and xAI are all racing to develop more capable systems while also addressing concerns around safety and responsible deployment. Anthropic's approach suggests that future AI releases may increasingly combine powerful capabilities with strict usage controls.

For users, Fable 5 represents access to a new generation of AI that can handle more complex tasks than previous models. For businesses, it offers a glimpse of how advanced AI could reshape software development, research, automation and decision-making over the next few years. Whether that transformation happens quickly or gradually, one thing is becoming clear: AI systems are moving beyond simple chatbots and becoming active tools capable of carrying out increasingly sophisticated work