Moonshot AI: the Chinese prodigy that Washington accuses of copying Anthropic

Fable 5, the latest large language model developed by Anthropic, was launched on June 9 before being abruptly removed just three days later. The decision came after U.S. Secretary of Commerce Howard Lutnick requested that Anthropic suspend access to the model, citing national security concerns. This move followed pressure from American authorities worried about the potential risks posed by such a powerful AI system in cybersecurity contexts. Anthropic complied with the request and restricted access to all users to strengthen its safety measures. After a three-week suspension, Fable 5 made a comeback at the beginning of July. Anthropic positioned the model as its most advanced version for complex reasoning tasks, though access remained limited to select users. To evaluate its capabilities, Le Figaro enlisted Hamza Bouanani, a practice manager specializing in AI projects at consulting firm Margo. Bouanani put Fable 5 through rigorous tests to compare it against other models and assess its unique strengths. During testing, Fable 5 demonstrated exceptional reasoning abilities, often solving problems without requiring explicit guidance. According to Bouanani’s evaluation, the model's capacity to understand context and generate accurate responses set it apart from competitors. It handled intricate queries with remarkable speed and precision, showcasing a level of sophistication that impressed even seasoned experts in the field. These results highlight how Fable 5 could potentially revolutionize industries reliant on advanced AI applications, from finance to healthcare. The deployment of Fable 5 had initially faced delays due to regulatory scrutiny. U.S. officials were concerned about the implications of deploying such a potent AI tool, particularly in areas involving sensitive data and critical infrastructure. The temporary withdrawal allowed Anthropic to refine its safeguards and ensure compliance with evolving standards. Now, with the model back online under stricter controls, the company aims to balance innovation with responsibility. Anthropic has not disclosed specific details about the nature of the security concerns raised by U.S. authorities. However, industry analysts suggest that the focus was likely on preventing misuse or unintended consequences that could compromise national interests. Despite these restrictions, Fable 5 continues to operate within a controlled environment, allowing selected users to benefit from its capabilities while minimizing potential risks. As the AI landscape evolves rapidly, Fable 5 represents a significant step forward in the development of highly capable language models. Its performance during testing underscores the growing importance of models that can reason independently and adapt to complex scenarios. While the broader public may not have direct access, the insights gained from its deployment will likely influence future advancements in AI technology. For now, Fable 5 stands as a testament to the ongoing efforts of companies like Anthropic to push the boundaries of artificial intelligence responsibly.