**Anthropic** has confirmed plans to make its Mythos-class models generally available after a delay caused by potential security vulnerabilities affecting public and private software. The company initially announced **Mythos** in April as a restricted model, granting access only to select organizations, including security researchers. Initially, **Anthropic** held back from a public release due to significant "security" concerns associated with the **Mythos** model. "The advantage will belong to the side that can get the most out of these tools," **Anthropic** warned back in April. "In the short term, this could be attackers, if frontier labs aren’t careful about how they release these models. In the long term, we expect it will be defenders who will more efficiently direct resources and use these models to fix bugs before new code ever ships." AI companies typically exercise caution when releasing powerful models, ensuring robust safeguards are in place to prevent misuse. It appears that **Anthropic** has successfully developed these necessary guardrails for the **Mythos** model, which is rumored to be substantially more capable than Opus 4.8 and other publicly available models. ## Anthropic Prepares Rollout of Mythos Model In a recent [blog post](https://www.anthropic.com/news/claude-opus-4-8), **Anthropic** stated its intention to release Mythos-class models to the public in the coming weeks, although a specific release date wasn't provided. "We’re making swift progress on developing these safeguards and expect to be able to bring Mythos-class models to all our customers in the coming weeks," the blog post stated. **Anthropic** is already allowing a limited number of organizations to use a preview version of **Claude Mythos** for cybersecurity applications, but it remains unclear whether the public release will be identical. According to the company, the **Mythos** model exhibits significant improvements in code reasoning and autonomy, surpassing the capabilities of **Claude**'s current flagship model, Opus 4.8. Notably, the "Mythos-preview" model briefly appeared on **Claude Code** for some users before being taken offline. <div><p><a rel="noopener nofollow" href="https://hubs.li/Q048zztN0"><img alt="article image" src="https://www.bleepstatic.com/c/p/validation-gap.jpg"></a></p><div><h2><a rel="noopener nofollow" href="https://hubs.li/Q048zztN0">The Validation Gap: Automated Pentesting Answers One Question. You Need Six.</a></h2><p>Automated pentesting tools deliver real value, but they were built to answer one question: can an attacker move through the network? They were not built to test whether your controls block threats, your detection rules fire, or your cloud configs hold.</p><p>This guide covers the 6 surfaces you actually need to validate.</p><p><a rel="noopener nofollow" href="https://hubs.li/Q048zztN0">Download Now</a></p></div></div>