Anthropic Confirms Mythos Capybara Tier, Limits Release To Cybersecurity Trials After Draft Leak

Anthropic confirmed that Claude Mythos exists and is tied to a new Capybara tier above Opus, with a deliberately slow, cybersecurity-focused rollout.

“Here's what next as Anthropic's most powerful AI model leaked via unsecured data cache A draft blog post left in an unsecured data cache revealed a new model tier called Capybara that Anthropic says is more capable than anything it has built, with the company flagging "unprecedented" cybersecurity risks”

@coindesk

Fortune reported that an unsecured data cache in Anthropic’s content management system exposed roughly 3,000 unpublished assets, including a draft post describing Mythos and Capybara.

@coindesk

Anthropic described Mythos as 'a step change' and 'the most capable we’ve built to date,' signaling a high-stakes model whose public release is being tightly controlled.

Non-Western outlets highlighted the risk angle, noting that the documents warned of 'unprecedented cybersecurity risks' as defenders prepare for faster, more capable attacks.

The model is being trialed by a small group of early-access customers with cybersecurity focus.

Access via the Claude API is expanding gradually to vetted users.

CoinDesk

Anthropic described the rollout as deliberate and security-focused, aiming to harden defenses before broader release.

Mythos is described as 'currently far ahead of any other AI model in cyber capabilities' with warning of a vulnerability-exploitation wave.

“AI company Anthropic has inadvertently revealed details of an upcoming model release, an exclusive CEO event, and other internal data, including images and PDFs, in what appears to be a significant security lapse”

Fortune

Analysts argued the risk leans toward offensive cyber use rather than displacing existing defenses.

The rollout’s cybersecurity emphasis is framed as a defensive measure to allow defenders to harden systems first.

Two blog variants surface with different names for the same model—Mythos and Capybara—highlighting naming ambiguity in the leak.

Analysts and outlets frame Capybara as an entirely new tier above Opus, not a simple Opus upgrade.

Futurism

The leak underscores a broader trajectory toward high-capability models that demand defender-first rollout.

“An accidental leak has now been officially confirmed by AI company Anthropic regarding its most powerful AI model yet”

Mashable