Model 2とは｜公開しない最上位モデル

"Model 2, which is somewhat more capable than Mythos 5. Our rough qualitative sense is that this model is a noticeable improvement on Mythos 5 for many tasks relevant to internal use but does not display a capability jump of the degree observed from Claude Opus 4.6 to Mythos Preview."（Section 1.4 Notes on coverage of unreleased models）— Anthropic 公式リスクレポートより

"Claude Mythos 5 and Model 2 are used heavily within Anthropic for coding, data generation, and other agentic use cases."（Table 1.2.A）／"We focus our analysis on Claude Mythos 5 and Model 2, our most capable and most commonly internally used models."（Section 2.3）— Anthropic 公式リスクレポートより

"We do not currently have plans to release this model externally, and have not run all of our typical suite of predeployment assessments, so we have somewhat lower confidence in our beliefs about its capabilities."（Section 1.4）／"As of the coverage date of this report, Model 2 had been through the pre-internal-deployment review process described in Section 2.18, but had not been used or evaluated as extensively as Mythos 5."（Section 2.3）— Anthropic 公式リスクレポートより

"For the corresponding review of Model 2, we additionally piloted a staged deployment process in which we first deployed the model on internal surfaces with stronger blocking controls against dangerous actions to gather more real-world usage data in a lower-risk setting before rolling it out for unrestricted internal deployment."（内部展開前レビューの記述箇所）— Anthropic 公式リスクレポートより

"Low (an increase from our previous assessment of "very low," in light of general increased uncertainty around recent incident disclosures related to model behavior in cybersecurity evaluations)."（Table 1.2.A / Overall risk assessment）／"Covered risk is therefore very low based on these claims, but in an abundance of caution we assess risk to be only low due to increased uncertainty."（Section 2.6 Claims and core argument）／"We did not observe any new or more concerning forms of misalignment during Model 2's internal deployment approval process than what has been discussed above for Mythos 5."（Model 2 の内部展開承認に関する記述）— Anthropic 公式リスクレポートより

"Finally, we note that the UK's AISI recently published a report on a cybersecurity evaluation involving Claude Mythos 5, during which the model attempted to complete an assignment in a setup where its normal safeguards were removed and it was deliberately given internet access. AISI reports that the models "engaged in sustained, potentially harmful activity directed at real people and organisations." This incident occurred after the coverage date of this report,"（Model 2 の内部展開承認と UK AISI 報告に触れた箇所）— Anthropic 公式リスクレポートより