ANTHROPIC UNLEASHES ITS MOST POWERFUL AI ON THE GENERAL PUBLIC FOR THE FIRST TIME — AND LOCKED THE DANGEROUS STUFF IN A VAULT Anthropic just handed the world its most capable AI model, and the timing could not be more unnerving. Claude Fable 5, released June 9, marks the first time the company has made a version of its Mythos architecture available to the general public. For months, Mythos existed only behind closed doors, accessible to a handful of vetted organizations managing critical infrastructure across 15 countries. Now anyone with an Anthropic subscription can access it, at least through June 22. The model excels at software engineering, complex analytics, and agentic tasks that require sustained reasoning. In benchmark testing by Hex, it was the first model to score 90% on complex analytical workloads. Vibe-coding platform Base44 says it one-shots full applications better than anything else on the market. Rakuten claims its autonomous operation capabilities essentially pay for themselves. But Anthropic built a wall around the most sensitive corners of the model. Responses touching cybersecurity, biology, chemistry, and distillation are hard-blocked and rerouted to the older Opus 4.8. More striking: Anthropic is now requiring a 30-day data retention window on all traffic, scrapping previous zero-retention agreements for enterprise clients. The company says the data will not be used for training but will be used to defend against novel jailbreaks. Price: $10 per million input tokens, $50 per million output. Double Opus 4.8. The model is powerful enough that the conversation about cost is already happening inside every enterprise that has seen the bill come in. Keywords: Claude Fable 5, Anthropic Mythos, AI model release, Anthropic public model