Anthropic releases Mythos-like AI model to the public, Claude Fable 5

Source Domain: www.cnbc.com

Anthropic announced the release of a new AI model, Claude Fable 5, following two months of limited access to its more advanced Mythos model, emphasizing safety and risk controls.
The company highlighted that Claude Fable 5 is designed to provide significant benefits while guarding against high-risk misuse, particularly in cybersecurity and biology, by using advanced classifier systems.
Anthropic’s new model showcases exceptional performance in software engineering and knowledge work tasks, outperforming its previous model, Claude Opus 4.8, in some benchmark tests.
Anthropic aims to balance technological advancements with safety, managing the “race to the top” by deploying safeguards that block responses to harmful requests.
The release of these new models comes as Anthropic moves toward a significant public offering (IPO) and experiencing explosive growth, with its revenue run rate skyrocketing from $10 billion to $47 billion over the past year.
Anthropic’s latest developments coincide with a competitive AI market, with rivals like OpenAI also preparing for IPOs and new funding rounds.
The new model, Claude Fable 5, comes with higher costs per token compared to its predecessor, but Anthropic argues it provides increased value and efficiency for enterprise customers and paying subscribers.

You may have missed