Artificial intelligence (AI) model developer Anthropic opened its most powerful version of technology to the public on the 9th, while imposing restrictions on its use in sensitive areas.
This model, called Fable 5, is the first widely available model in Anthropic's new Mythos series; Mythos is Anthropic's most advanced AI product line, unveiled earlier this April, but was restricted due to cybersecurity concerns.
Anthropic stated that Fable 5 excels at writing and debugging software code, answering complex research questions, and analyzing images.
At the same time, Anthropic is providing an unrestricted version—Claude Mythos 5—to companies and organizations that already have access to this series of models, including cybersecurity partners who have joined the "Project Glasswing."
According to media reports, in April Anthropic announced that thousands of software vulnerabilities had been discovered through Mythos. So far, under the "Project Swallowtail" initiative, access has been limited to about 200 organizations— including the U.S. government. This selected group was expanded in early June to about 200 organizations across more than 15 countries, with further increases expected in the future.
Given Mythos’s ability to quickly identify vulnerabilities in critical infrastructure such as banking platforms and power grids, Anthropic has restricted access to this model due to cybersecurity considerations.
Anthropic stated that most security-related or bio-chemical queries will instead be routed to the lower-tier model Opus 4.8, which was made public at the end of May.
Anthropic pointed out that there have already been attempts to extract the relevant technology en masse for training competing AI models in authoritarian countries; such queries will likewise be redirected to the less capable model.